1

Religion and Public Goods Provision: Experimental and Interview Evidence from

Catholicism and Islam

Carolyn M. Warner, Arizona State University

Ramazan Kilinc, University of Nebraska Omaha

Christopher W. Hale, Northern Arizona University

Adam B. Cohen, Arizona State University

Kathryn A. Johnson, Arizona State University

Corresponding Author: Carolyn Warner, cwarner@asu.edu

DRAFT. 9.11.2013

Comments welcome, especially those with an eye towards helping us improve the paper for

publication.

This paper is a revised version of the 2013 Sage Award of the American Political Science

Association Qualitative and Multi-method Research Section for best paper developing or

applying qualitative methods presented at the preceding APSA conference.

2

Abstract

Religions such as Catholicism and Islam are generators of substantial amounts of charitable

donations and volunteer work, and they sustain themselves as organizations. How do they

produce charitable public goods and their own religious club goods when they are open to

extensive free-riding? We argue that mainstream religions facilitate club and public goods

provision by using their community structures and their theological belief systems to activate the

pro-social, generous tendencies of their members. The study is based on in-field essay-prime

experiments with over 800 Catholics and Muslims in Dublin and Istanbul and on case study-

based interviews with over 200 Catholics and Muslims in Dublin, Istanbul, Milan and Paris. Our

findings highlight the role of the religious concepts of duty to God, God’s grace, and

deservingness on adherents of the two religions, as well as religious community, though not in

the way typically expected by rational choice theory. The evidence shows support for

understanding individuals as being pro-social, not just strictly self-interested utility maximizers.

Our argument about the role of beliefs and community help explain why mainstream religions

can produce club and public goods despite their lack of strong monitoring and sanctioning

mechanisms. The paper also demonstrates the advantages of using a methodology that combines

field experiments with case study-based interviews.

3

Introduction1

From Indonesian tsunami relief efforts to health clinics in the Gaza Strip to promoting

civic culture in America, organized religions have been credited with providing public goods

(Davis and Robinson 2012; Dhingra and Becker 2001; Loveland, et al. 2005; Caputo 2009;

Iannaccone and Berman 2006; Trejo 2009, 323; Wiktorowicz 2004). How they do so is far less

understood than the fact that they do. Public choice approaches have made progress in

understanding why and how strict religious sects and terrorist groups, as well as ethno-religious

groups, provide public goods (Alexander and Christia 2011; Berman 2009; Berman and Laitin

2008 Cammet and Issar 2010). This literature has emphasized the collective action dilemma of

free-riding and focused on how groups monitor and sanction their members (Tsai 2007). Another

literature has focused on religious community affiliation and values, exploring whether different

denominations or whether the religious versus the secular are more or less likely to be charitable

or socially engaged (Sarkissian 2012). Some have found that the religious are “better neighbors”

than the non-religious because of their involvement in a religious community (Putnam and

Campbell 2010, 471).

In both streams of scholarship, the causal mechanisms leading mainstream religious

believers to provide public goods are unclear. Because mainstream religions often lack effective

monitoring and sanctioning structures, the expectation is that they would do a poor job of

creating collective goods (Makowsky 2011; Iannaccone 1994). Yet evidence suggests otherwise.

Religions such as Catholicism and Islam are generators of substantial amounts of charitable

donations and volunteer work, and they sustain themselves as organizations. How do they

produce charitable public goods and their own religious “club” goods when they are open to

extensive free-riding? Given the prevalence of religion in human society, and that Islam and

Catholicism claim over 2.5 billion followers combined, the question is important to study. We

re-frame this question to ask whether there are spiritual mechanisms that enable religious

communities to produce club and public goods. Mainstream religions may facilitate club and

public goods provision by using their theological belief systems and communities to activate the

pro-social tendencies of their members. We test this possibility with data collected from

experiments and interviews with Muslims and Catholics in Europe.

We focus on Italy, France, Ireland and Turkey. Italy, France and Ireland have been and

remain crucial to the history and life of the Catholic Church and Catholicism; Turkey is a major

Islamic country, with a significant role in Islam’s history. Ireland, Italy and France are

predominantly Catholic, and each has Muslim populations in their major cities (Maréchal et al.

2003; Sakaranaho 2006, 270-314). Turkey is predominantly Muslim and has a small Catholic

minority (Bailey and Bailey 2003, 196-7; Frazee 1983, 223-238; Le Coz 1995, 381; Tcholakian

1998, 387-412). To limit confounding variables, we are restricting our project to stable

democratic countries in Europe. With most of Europe’s populations living in urban areas, we

located our research in large urban centers of Paris, Milan, Dublin and Istanbul, studying a

1 We thank Sevde Arpaci, Matthew E. Bergman, Angela Pirlott, Scott Swagerty and Chloe Westlund for research

assistance. We thank the University of Notre Dame Science of Generosity Program, and the Institute for Social

Science Research at Arizona State University for their financial support of this research and the Arizona State

University Center for the Study of Religion and Conflict for logistical support. We are very grateful for the

assistance and cooperation of the religious officials who allowed access to their religious communities. Professors

David Farrell and Christopher Whelan at University College Dublin and Professors Ali Tekcan and Kutlu Űlgen at

Boğaziçi University, Istanbul provided invaluable help in facilitating the conduct of experiments with students at

their universities. We especially thank the interviewees and experiment participants for their willingness to take part

in this project.

4

Catholic parish and Muslim organization in each.2 These four cities enable us to study public

goods production of Catholicism when in majority and minority status, and of Islam when in

minority and majority status.

We use data collected from experiments conducted with over 800 students and

community members in Dublin, Ireland and in Istanbul, Turkey and from over 200 interviews

with Catholic parishioners, Muslim association members and religious officials in Dublin,

Istanbul, Milan and Paris.3 This multi-methods approach combines the advantages of

experiments and in-depth interviews and case studies. Experiments randomly manipulate

treatments across subjects and allow for quantitatively measurable differences across treatment

groups. In-depth interviews within case studies enable us to probe understandings of concepts

and motives, observe behaviors, and learn directly about organizational structures (Gerring 2007;

Wuthnow 2011). This mitigates the disadvantages of either approach separately: experiments are

open to the charge of being artificial, lacking external validity, and hampered by hard to

anticipate design problems (Henrich, Heine and Norenzayan 2010). Interviews and case studies

might not be “clean” enough to establish causality because subjects are not randomly selected,

treatments are not manipulated (though they are compared across groups), and interviews only

tell us what people think motivated them, not what “really” motivated them (Chaves 2010).

Using data from both methods to cross-check findings from each method should generate more

confidence in our conclusions.

Our results suggest that several religious beliefs and community dynamics influence the

propensities of individuals to engage in the production of club and public goods, and these

effects vary across the two religions. This coheres with findings from psychology and behavioral

economics that humans are motivated by pro-social orientations and not just by avoidance of

harm (Johnson, Li, Cohen, & Okun, 2013). Yet the results, intriguingly, were also somewhat

different between the experiments and the interview-based case studies, and we will discuss

likely reasons why in a later section. In the experiments, Catholics were much more likely than

Muslims to exhibit generous behavior in response to a prompt of God’s grace, duty to God, or

teachings of the religion about deservedness of those in need. Our interviews told a somewhat

different story. Catholics overwhelmingly denied having a “duty to God” to be helpful or

generous; Muslims overwhelmingly said they did have such a duty. Catholics and Muslims

indicated they are influenced by their religious communities, though less in the way rational

choice theories of public goods expect and more in the way theories of pro-sociality expect.

Catholics and Muslims helped members of their respective in- and out-groups, usually by

donating to and volunteering in one of their in-group organizations. In the experiments Catholics

were more likely to give to an out-group than an in-group, and Muslims more likely to do the

reverse. The interviews revealed that both religions are oriented towards helping others, of in-

and out-groups, often through organizations of the in-group. We comment on this discrepancy in

the discussion. The dual methodology of experiments and case study-based interviews shows the

advantages of applying both approaches to our study. Each helps check the findings of the other,

and enables us to avoid reaching erroneous conclusions in light of apparent non-findings from

2 We did not choose the capital cities of Italy or Turkey because of the need to have an adequate population of the

religious minority in question. There are very few Turkish Muslims in Rome, and very few Roman Catholics in

Ankara. Milan and Istanbul had requisite minority populations. 3 Throughout the paper, when we mean the local institutions, we use Gülen “association”. When we mean the entire

organization at the local, national and international levels, we use Gülen “movement.” This is in keeping with how

we use “parish” and “church” for the local level study of Catholics, and “Catholic Church” for the entire

organization of the Catholic Church.

5

one or the other method, and finally, enables us to refine the methods and questions for future

research.

First, we present the current state of the literature assessing the relationship between

religious organizations and the production of club and public goods. We suggest that previous

studies have paid insufficient attention to the micro-level mechanisms through which religions

prompt individuals to contribute to both. We then discuss our theoretical framework. It builds on

a multidisciplinary body of literature arguing that rational calculations of utility fail to recognize

that individuals have pro-social orientations as well as self-interested ones. We suggest that

religious beliefs and communities influence the degree and form that pro-social orientations take

in individuals. We then present our research design, findings from the experiments and

interviews, and a discussion of results. We conclude with a restatement of the main themes and

findings of the study.

Background Literature

One stream of literature in political science has pointed to religion’s role in producing

“social networks, shared solidarities, and leadership structures” that facilitate collective action

(Tarrow 1994, 11). Other work relies on factors influencing religious elite cost/benefit

calculations that prompt religious leaders to support particular social movements (Gill 1998;

Trejo 2009). Neither has addressed the question of how an organized religion or set of religious

beliefs might prompt adherents to participate in the production of collective goods despite having

an incentive to free-ride off the efforts of others.4

A related literature on ethnicity and public goods provision points towards incentives and

sanctions, and the mobilizational capacities of group structures (e.g., Alesina, Baqir, and Easterly

1999; Fearon and Laitin 1996; Habyarimana, Humphreys, Posner, and Weinstein 2007). Within

political science, this perspective has not been widely applied to the study of religious groups

and collective goods provision. Economists have turned their attention to this issue, developing a

theory of religious clubs. The club goods approach reasons that clubs provide their members

with jointly produced excludable goods that are subject to overcrowding and congestion if the

size of the group becomes too large. Clubs accordingly enact mechanisms to restrict membership

and exclude others from the consumption of the goods provided by the club (Samuelson 1954;

Buchanan 1965; Cornes and Sandler 1996). Applied to religious organizations, religious clubs

use sacrifice and commitment requirements to exclude free-riders and restrict the provision of

religious goods to committed members (Iannaccone 1992). Religions that enact strict

membership requirements are more efficient and effective "club" good providers. This

framework has been expanded and empirically evaluated in a variety of settings (McBride 2007;

Chamlee-Wright and Storr 2009; Chen 2010; Orman 2011; Makowsky 2011), and has most

notably been used to explain the seeming comparative advantage of religious sects in

perpetrating terrorist acts (Berman 2009; Berman and Laitin 2008; Iannaccone and Berman

2006).

Club and public goods differ on the matter of excludability. Non-participation in the

production of a club good can be sanctioned by withholding various club-goods from non-

contributors. Public goods, on the other hand, have no such possibility of exclusion. This

exclusionary principle has important ramifications. Exclusion to weed out non-contributors

4Collective goods are goods provided by multiple individuals, of which public goods and public goods are both

subsets. Public goods are non-excludable and non-rivalrous; club goods (within a club) are excludable and non-

rivalrous, though there may be a “menu” of club levels and excludability (McBride 2007).

6

explains why individuals would contribute to the production of "club" goods in a religious

context. A religious club could exclude members from access to a variety of desired club goods

if members shirk from their productive responsibilities. Exclusion from the club does not explain

why religion would cause individuals to contribute to the production of "public" goods, whose

benefits do not accrue to the particular organization, and whose enjoyment cannot be restricted to

contributing members.5

In the religious economics framework, collective outcomes result only from the “self-

interested decisions of strictly rational actors” (Iannaccone 1994, 1185). While the framework

provides leverage in explaining how and why strict sects efficiently provide club goods to their

members, it is less successful in explaining how and why mainstream institutions, with

substantially lower sacrifice and commitment requirements, provide club or public goods.

Religious groups also serve out-group members as well (Cammett and Issar 2010). The question

of why some within the group give and volunteer to provide club and public goods, and allow

others to free-ride, is unanswered.6

There is a sustained empirical challenge to conventional rational choice analysis

concerning whether individuals consistently work on behalf of advancing their own material self-

interest. Experimental evidence suggests individuals are socially oriented and have preferences

to engage in pro-social behavior. Sanctions and incentives are not the only or necessary factors to

bring about generous, cooperative behavior in individuals (Andreoni and Rao 2011; Warneken et

al 2006; Warneken and Tomasello 2008; Warneken and Tomasello 2009). The distribution of

these considerations is not uniformly distributed across the population (e.g., Andreoni 1990;

Andreoni and Miller 2002; Bowles and Gintis 2004; Casari and Plott 2003; Fischbacher,

Gächter, and Fehr 2001; Kurzban and Houser 2005; Simpson and Willer 2008). The literature

has turned to relying upon some level of altruism or “warm glow” rewards of volunteering and

giving to explain the existence of public goods in moderate religions (Hungerman 2009). There

is little theorizing about the conditions under which individuals might be more or less inclined to

display altruistic social preferences. Instead, the distribution of these altruistic tendencies is

theorized as exogenous and idiosyncratic individual-level traits. We propose that religious beliefs

and community attachment may help explain the distribution of altruistic behaviors in some of

the population.

Because religious teachings typically encourage the faithful to subordinate their own

desires to help others, religion could increase pro-social tendencies (Norenzayan and Shariff,

2008; Shariff and Norenzayan, 2007; cf. McBrien 1980, 977). In Christianity, New Testament

5Previous work on the relationship between religious sects and public goods is particularly challenged by

this point (Iannaccone and Berman 2006; Berman and Laitin 2008). The positive strategic and political externalities

derived from terrorist attacks cannot be confined simply to terrorist or religious groups, but instead accrue to the

surrounding society at large. Terrorist groups such as Hamas, for example, often do not restrict social welfare

programs to their members, but instead make them broadly available to society (Levitt and Ross 2006, 108-109).

Even if the terrorist group has an interest in providing these services in the interest of generating support for the

organization from the local population, it is not at all clear from a micro-level perspective why adherents to the sect

are devoting time and energy to produce public goods they can presumably receive whether or not they devote the

time and money necessary for their production. 6 Makowsky, using an agent based model, attempts to explain individuals’ contributions by their wage-based

opportunity costs (Makowsky 2011a). To our knowledge, Makowsky’s argument has not been tested empirically. In

strict rational choice theory, Makowsky’s argument might work, but it requires that each congregant have a

significant amount of accurate information about each other congregant’s income and work/retirement status. This

world of complete information does not obtain empirically.

7

texts “suggest strongly the centrality of giving and service to the religious life,” as do other

Catholic Church teachings (Queen 1996, 27; Catechism 1999, 461) Islam has several explicitly

described institutions that call for charity; the most well-known is the obligatory zakat (alms-

giving), one of the five pillars of Islam (Clark 2004, 8; Benthall and Bellion-Jourdan 2003, 7-44;

Kozlowski 1998; Al-Ghazzali 1966). Thus, organized religions could have a big role in turning

“on” and channeling the pro-social nature of individuals. In this paper, we test the possibility that

religions can systematically elicit the pro-social tendencies of individuals through their

theologies and community practices.

Theory and Hypotheses:

We postulate that variations in theological distinctions and community practices impact

the weight of pro-social orientations on an individual's behavioral considerations, as expressed

through their contributions to the production of either a club or public good. 7

We use the giving

of ones time and other resources to others, often termed “generosity”, as a proxy for club and

public goods (see below). Religion is a multi-faceted phenomenon with different social effects,

depending on what the construct of interest is (Preston, Ritter and Hernandez 2010). We distill

several major features of religions to assess the impact of specific beliefs on religious

individuals’ propensity to donate to club or public goods. We further differentiate expectations

for Catholicism and Islam.

A variety of disciplines converge on the argument that humans, as social beings, are

attuned to what their co-equals think of them and adjust their behavior accordingly. Groups

affect individual behavior partly through norms and expectations. Applied to the concept of pro-

social helping behavior, this perspective interprets generous acts as those required and rewarded

by the group. Pro-sociality is generalized group cooperation: the individual contributes because

of group expectations and fear of group sanctions if the expectations are not met (Mauss

1954/1967; Peterson 1993; Sahlins 1972; Tsai 2007). Religions create groups, those groups have

a set of expectations about what is moral, or proper, behavior (Graham and Haidt 2010, 143-5;

Durkheim 1915, 44). As do other religions, Catholicism and Islam both create religious

communities, and while each religion has a large world-wide community, each also has a local

and small community in which most adherents practice their faith and interact with others of

their faith. Islam and Catholicism consider integration into and cooperation with a community to

be an inherently valued way of serving God. Because both Catholicism and Islam are theorized

to be religiously collectivist, and to promote charity within the community (Cohen et al. 2005;

Cohen and Hill, 2007; Cohen and Rozin 2001; Cohen, Siegel, and Rozin 2003; Kniss and

Numrich 2007, 56-7), beliefs about the normative expectations of the religion should lead to pro-

social, cooperative behavior for the group. These are religious beliefs about how people should

act.

Hypothesis 1: Beliefs about the normative expectations of the religious community lead

Catholics and Muslims to contribute to club goods.

7 There is an extensive debate in social psychology on whether altruism or pro-social orientations that are not

ultimately self-serving exist (Cialdini et al 1997; Batson 2011). We recognize the debate, and here suggest that

rather than invoking a randomly distributed “warm glow” effect, as behavioral economists have done, to explain the

unexpected helping behavior of some individuals, we can see if particular aspects of religions systematically trigger

such behavior.

8

Religions might heighten self-other identification, making individuals empathetic and

more likely to aid others. Individuals are then charitable to others, because they can see

themselves in the other’s situation. The most basic element of social identity theory is that

people's self-definitions are affected by their membership in social groups along with the value

they place on and emotional significance of those memberships (Tajfel and Turner 1979; Abrams

et al. 1990). They more easily recognize themselves in others (empathy) who are in the in-group

rather than in the out-group (Cialdini et al. 1997). Identification in their religious group would

lead Catholics and Muslims to prioritize helping their in-group. However, because both religions

emphasize a universality of humankind, Catholics and Muslims should not prioritize their in-

group. Instead they would be empathetic with others, no matter what their group affiliation.

Therefore, we expect the following:

Hypothesis 2: Self-other identification fostered by their religion makes Catholics and

Muslims more likely to contribute to the production of public goods. The effect on the production

of club goods will be less pronounced.

Religious discourse may emphasize additional other-regarding attitudes and behaviors

such as deservedness of someone in need. Previous research has found that adherents of different

religions vary in their view of whether a needy person is deserving, and the more they view the

target as deserving, the more helpful they are (Will and Cochran 1995). Catholicism and Islam

both have as key tenets of their religions that the poor and needy are deserving of help regardless

of the circumstances, character and behavior of the needy individual (Queen 1996; Tropman

2002). From the perspective of the potential impact of their religious beliefs, we would therefore

expect Catholics and Muslims to exhibit more helpful behavior when their thinking about

someone in need is couched in terms of that person’s deservingness, and that it would not be

subject to an in-group bias.8 .

Hypothesis 3: The concept of deservedness promoted by their respective religions should

lead Catholics and Muslims to be more generous towards others, no matter the others’ religion.

Many religions emphasize obedience to a divine being, and Catholicism and Islam would

seem to be no exception. What constitutes obedience may vary across religions. In Catholicism

and Islam, obedience might be a duty to obey God’s will and commandments. As Pope Benedict

XVI writes, “Each person finds his good by adherence to God’s plan for him, in order to realize

it fully” (2009, 1). The word “Islam” means “voluntary submission to God,” sometimes also

translated as “submission to the will of God.” We operationalize the concept of obedience to

God’s will as “duty to God”--an obligation to obey God. If God’s commandment is that the

individual be other-regarding, then religiosity affects helping behavior by increasing Catholics

and Muslims’ sense of an obligation to God. Both religions emphasize helping others, though

there may be differences in whether the helping is an obligation. Islam requires, through zakat

and fitr,giving resources to others.9 While Catholicism speaks of tithing, tithing is not mentioned

as an obligation, nor is it a formal ritual of the theology.

Hypothesis 4: The concept of duty to God should make Catholics and Muslims more

generous towards others, contributing to the production of both club and public goods. This

effect will be more pronounced for Muslims than for Catholics.

8 In the absence of religious commandments to help the poor, one might instead expect the in-group to be favored

(Hewstone, Rubin and Willis 2002) 9 Islam also emphasizes non-obligatory giving, or sadaqah (Queen 1996, 49-50).

9

Helping others could also be affected by the divine inspiration to live a life filled with

love in Catholicism (Benedict XVI 2009, 1), or a life lived as the prophet Mohammed lived in

Islam. A spirit of generosity may be born from that faith in God brings freedom to give

abundantly. Religion may engender generosity through spirituality and a sense of grace, a

commitment to and faith in God, God working “through, with, and in us” (Himes 2006, 18, 19;

Benedict XVI 2009; Kozlowski 1998, 282; Queen 1996, 27, 50). Catholicism specifically

expects “faith, hope and charity” to be the expression of God dwelling within the person

(McBrien 1980, 991). Pope Benedict XVI makes the connection explicit, writing, “Charity is

love received and given. It is ‘grace’ (cháris)” (2009, 3). In Islam, giving stems from being a

“righteous man” or woman. It is “for the love of God” that one gives to others (Qur’an, Sura

2:177). Both religions have an understanding of God’s grace, that it is through His grace and

mercy that one enters heaven/paradise, and that one loves others. God’s grace may be a

mechanism for prompting pro-social behavior.

Hypothesis 5: The inspiration of God’s grace that is fostered by their religion makes

Catholics and Muslims more likely to contribute to the production of both club and public goods.

Conventional wisdom in social sciences holds that religion generally may prompt pro-

social behavior . While our view is that specific beliefs and social norms within a religion

prompt such behavior, and that “religion” is such a multi-faceted concept that it is not clear what

construct “religion” would be priming, we acknowledge the extensive body of research (which

gives conflicting results) showing that religion prompts generosity.10

We refine the hypothesis in

keeping with some studies of prejudice and religion arguing that religion heightens in-group

identification at the expense of out-group identification.

Hypothesis 6: The concept of religion will make both Catholics and Muslims to

contribute to collective goods, and more likely to contribute to club than to public goods.

Methods and Data

To test our hypotheses, experiments were conducted in both Ireland and Turkey in two

waves, one in the fall of 2010, and the second in the summer of 2011 and interview-based case

studies were conducted in various months from May 2010 to May 2011, with one researcher

spending approximately four weeks with a religious group.11

Experiments

The experiments were designed to assess the ways in which our hypothesized mechanisms

impact the propensity of experimental subjects to make a charitable donation to either their

religious in-group (conceptualized as a club goods donation), or to make a donation to a secular

group charity (conceptualized as a public goods donation). The experiments used primes of the

focal constructs. They were not structured to see how chronically salient ideas of duty to God, or

community, etc., relate to generosity, but whether priming these resulted in generosity. Four

separate experimental samples were taken in the first wave. Within Ireland, adult Catholic parish

10

Preston and Ritter (2013 in press) argue that religion is understood as a principle that defines a group, and that

people distinguish it from cognitive beliefs about supernatural being(s) or edicts of that religion. Their research is

based on experiments with US university students priming “religion” and “God”. 11

Kilinc conducted the research on the Gülen movement associations in Istanbul, Paris, Milan and Dublin, and on

the Catholic parish in Istanbul; Warner conducted the research on the Catholic parishes in Paris, Milan and Dublin,

and Cohen conducted experiments on university students and community members in Istanbul and Dublin.

10

members were recruited to take part in the experiment in Dublin, while undergraduate university

students were recruited to participate in the experiments at University College Dublin. In

Turkey, adult Muslim Gülen Association members were recruited to take part in the experiment

in Istanbul, while undergraduate university students were recruited to take part in the experiment

at Boğaziçi University in Istanbul.12

In the second wave, two separate samples were taken. One

sample involved adult Catholic parish members in Dublin, while the other involved adult Muslim

Gülen association members in Istanbul. We refer to the adult Catholic parish and Muslim

association members as “community members.” They were offered 25 Euros, or 25 Turkish Lira,

respectively, to participate in the study, while university students were offered 10 Euros or 10

Turkish Lira, respectively, to participate.13

With university students, we control a number of

confounds such as education level and socioeconomic status (Cohen 2009). But, university

students may mute religious differences and life cycle effects because they are more educated

and secularized than the general population, and less experienced (Rozin 2003; Henrich, Heine

and Norenzayan 2010). To gain a better representation of the population at large, we also

recruited general community participants. 14

Experiment participants were recruited through religious organizations (Catholic

churches in the Dublin area in Ireland, Gülen centers in and around Istanbul) and University

College Dublin (Dublin) and Boğaziçi University (Istanbul) and run in small groups (n’s of about

15-30) at their religious centers and university classrooms, respectively. Participants were seated

with room between them so as to preserve anonymity.

In the experiments, Muslims and Catholics were randomly assigned to one of six

treatments to prime theoretically derived causes of religiously motivated giving to others, or

were assigned to one control group (cf. Nickerson 2005, 235). Generally speaking, our approach

made salient to participants, via experimental priming manipulations, one of six theoretically

derived causes of generosity. Priming experiments make a certain concept or psychological

process salient (Bargh and Chartrand 2000), here a theorized specific causal mechanism of other-

regarding behavior. Certain experimentally validated ways of priming religion are already in the

psychology literature. For example, Shariff and Norenzayan (2007) subtly primed religion via

sentence completion tasks and documented that this promoted more generous behavior in

experimental economics tasks. However, because of the general nature of this prime it is

impossible to know what precise aspect of religion drove the increase in other-regarding

behavior. A recent paper by Preston and Ritter (2011) finds that there are differences when one

12

It is not possible to identify an Islamic movement that all Muslims and scholars would agree is representative and

typical of Islam. However, Islam’s basic message of charity is constant across the religion, and the Gülen Movement

has several advantages for our study: it is a relatively mainstream transnational organization within Islam and it is

relatively transparent and open about its activities, it is present throughout Turkey, and we have good contacts with

the organization (Yavuz 2003, 133-206; Kilinc 2007; Kuru 2003; Kuru 2005). To accommodate the fact that in some

Muslim dominant countries, mosques are used only for prayer and are state run (Turkey) or state supported, our

study of Turkey focuses on a Gülen organization rather than on a mosque. 13

These amounts were based on 1) the exchange rate of 1 euro=1.97 TRY (Turkish lira) in July 2010; 2) purchasing

power parity measured by the Economist Big Mac Index 2010 at http://bigmacindex.org/2010-big-mac-index.html.

This showed 5.95 TRY = 1 Big Mac and 3.3 Euro = 1 Big Mac (that, though, is for the entire eurozone); and 3)

discussions with faculty in Dublin and Istanbul who had conducted psychology experiments with university

students, and on those faculty members’ assessments of appropriate sums for community members. 14

Inchicore Parish; St. Bridged’s Church in Cabinteely; Knocklyon Parish; and St. Josephine at River Valley in

Swords; the Ihsander community associations in Istanbul. The groups recruited from did not contain those who were

interviewed.

11

primes subjects with “God” versus “religion”. Here we attempt to isolate specific causal

mechanism by priming duty to God, community expectations, self-other overlap, deservedness,

or religion. In this experimental work, we measure club and public goods provision in a

behavioral way, by gauging actual donations to charities (club or publich). By doing this,

priming experiments can test the specific concepts or processes that are causally

connected. Priming was done by asking participants to write essays about the specific concept,

such as duty to God. Within all experimental samples, subjects were randomly assigned to

receive one of seven experimental conditions.

The experiment was entirely questionnaire based. First, participants read a page

describing their rights as research participants. Then they were asked to write an essay, which

was our priming condition. Participants were primed by 1 of 7 levels: writing about their desk

and chair—our control condition; religion in general—a general religious control condition; duty

to God; God’s grace; religious community; religious teachings regarding self-other overlap, or

similarity to others; and religious teachings regarding others deserving help. Participants were

asked to write or list their thoughts and feelings about the topic. In order to better isolate the

effects of the hypothesized focal constructs, we did not set participants in strategic game

contexts.

Next, participants were asked to allocate their experimental payment (25 TRY or 25 Euro

for community members; 10 TRY or 10 Euro for university students) between keeping it for

themselves, donating some or all to an in-group charity (Catholic Children’s Fund or Islamic

Children’s fund) or donating some or all to an out-group secular charity (United Nations

Children’s Fund, UNICEF). Participants were told they could divide the payment any way they

wished. This was our major dependent variable, a behavioral measure of generosity. To make

these choices as anonymous and demand free as possible, participants were paid individually by

a second experimenter in a separate room. Finally, participants completed self-report scales of

intrinsic and extrinsic religiosity, and demographics (sex, marital status, educational attainment,

age, occupation, social class). The experiment took about 20-30 minutes for most participants.

At the end, participants were given their entire payment to keep and debriefed.

The contribution to either the Catholic or Islamic charity is conceptualized as a

contribution to a club good. While respondents are assured that these in-group charities do not

discriminate in the provision of charity on the basis of religion, the implication is that the charity

is administered by fellow members of the religious in-group. Any reputational benefits received

by administering the charity is confined to the in-group, as are proselytizing opportunities.

UNICEF, on the other hand, represents a pure public good. The benefits of making a contribution

to the charity cannot be confined to in-group population, but instead manifest themselves outside

the group.

The control condition had nothing to do with religion. Comparing any of the conditions

with the control will tell us if the prime influences generosity. The dependent variable is

conceptualized dichotomously, and in three distinct forms, assessing 1) whether individuals

made any donation at all, regardless of whether to the in-group or the out-group; 2) whether

individuals made a donation to the religious in-group; and 3) whether individuals made a

donation to the religious out-group.15

15

Most of the participants kept their entire participant payment or did not donate to one of the specified donations

groups (n=547). A sizeable minority (n=172) donated their entire participant payment. Ninety donated only a

portion of their participant payment, while keeping the rest for themselves. This characterized 11.1% of the

experimental population, and 39.1% of the individuals who gave a donation. If conceptualized continuously, the

12

Results

Table 1a indicates the total number of donations made across the experimental

population, and 28.4% of participants made a donation.16

Overall, participants were more likely

to have given to a religious out-group than to a religious in-group: 15.0% of respondents gave to

a religious in-group while 22.3% of respondents gave to the religious out-group.

Table 1b point out some interesting differences between the Catholic and Muslim

samples. The first is that Catholics were more likely to make a donation than Muslims: 39.2% of

the Catholic participants made some form of donation, compared to 19.8% of Muslim

participants. Catholics were less likely to have made a donation to the in-group (16.6%), and

more likely to have made a donation to the out-group (31.5%). The Muslim sample, on the other

hand, was more likely to have made a donation to the religious in-group (16.8%) than to the out-

group (10.8%).17

[Tables 1a and 1b about here, see p. 26]

The experimental conditions (Community, SelfOther, Deservedness, Duty, Grace, General

Religion, General Control) are included as categorical variables. The general control variable,

which is the “chair” essay prompt condition, has been excluded from the regression equation,

and thus odds ratios for each of the experimental conditions are interpreted in comparison to the

control condition. “Sample” is a dichotomous variable indicating whether an individual was part

of the community or student sample. Separate analyses are run for the Catholic and Muslim

populations.18

The community and student samples are pooled into each of the Catholic and

Muslim analyses. Logit analysis was used to analyze experimental data given the dichotomous

nature of the dependent variable and given the ability of maximum likelihood estimation to

control for the potential differences across sampled groups.

Table 2 presents results from the logit regression assessing whether Catholic or Muslim

participants made any donation, be it to the religious in-group or to the out-group.

[Table 2 about here, see p. 27]

For Irish Catholics, Duty to God exhibits statistically significant effects. Irish Catholics

who received this prime are 2.6 times more likely to give than individuals in the control group

For Turkish Muslims, none of the primes exhibit statistically significant effects. The sampled

group has statistically significant effects in both analyses. In the Irish Catholic analysis, students

are 70% less likely to have made a donation than community members, while students in the

Turkish Muslim sample are in fact nearly 3 times more likely to make a donation than

individuals in the community sample.

[Table 3 about here, see p. 28]

resultant U-Shaped distribution violates the assumption of normality and is not amenable to OLS analysis. Running

data in an ordinal logistic regression, while coding the donation variables with categories of no donations, partial

donations, or full donations, does not have a marked impact on model results. 16

Figures are rounded to first decimal point. In tables they are presented to second. 17

All of the above simple differences are statistically significant at least at the 95% confidence interval. 18

Those in the Irish pool indicating they were not Catholic were not included in the analyses of Catholics; those in

the Turkish pool indicating they were not Muslim were not included in the analyses of Muslims. See tables 5&6 for

basic demographics.

13

Table 3 presents results for whether Catholic and Muslim participants made donations

specifically to the religious in-group. There are no statistically significant effects by the primes

in either the Irish Catholic or Turkish Muslim groups. The Sample group exhibits similar effects

on the propensity to give to ingroup as it did to the inclination to give generally. Irish Catholic

students are 66% less likely to make a donation than their community counterparts, while

Turkish Muslim students are 2.6 times more likely to make a donation to the religious ingroup

than community members.

[Table 4 about here, see p. 29]

Table 4 presents the results of the odds that individuals in the Catholic and Muslim

samples gave to the out-group. The odds individuals who received the Deservedness, Duty to

God, and Grace primes gave to the out-group were increased by factors of 2.7, 2.9, and 3.2

respectively over the control group. Like the rest of the analyses, the primes had no effect on the

Muslim population. Irish Students are 61% less likely to give than community members, while

Muslim students are 5.9 times more likely to give than Muslim community members.

Discussion of the Experiment Results

A clear result of this analysis is that the primes had differential effects across the Catholic

and Muslim samples. While Catholics tended to respond to the Duty to God, Grace, and

Deservedness primes, priming these belief constructs did not have any effect on Muslims’

inclinations to give.

Regarding Hypothesis 1, we find little evidence of religious Community Expectations

having an effect on believers. Within both populations, Community Expectations was not found

to make individuals more likely to contribute to the production of club or public goods.

Hypothesis 2, that Self-other Overlap identification fostered by the religion’s teachings

makes individuals more likely to contribute to the production of club goods, but not public goods

for Muslims finds no support in our analysis. It also had no effect on Catholics.

For Hypothesis 3, the results from our analysis on the effect of the deservedness prime

suggest that it had a conclusive impact on out-group donations amongst Catholics. Catholics did

not find their own to be more deserving. There was no impact on Muslims.

Hypothesis 4, that religious organizations foster a sense of a “Duty to God” which then

prompts donations to both club and public goods, found strong support with regard to the

Catholic sample and no support among the Muslim sample. The Duty to God prime was the

single most influential prime amongst the Catholic population, with statistically significant

impacts in the tendency to make any donation and out-group donations. It generally had one of

the largest effect sizes among the Catholic population in the various analyses. In contrast, the

prime seems to have no effect in the Muslim population in any of the analyses.

Hypothesis 5, that the inspiration of God’s Grace prompts both club goods and public

goods donations, finds strong support in the Catholic sample and no support in the Muslim

sample. God’s Grace displays a strong effect, an odds ratio of 3.18, with regard to whether out-

group donations are made in the Catholic population. It has no statistically significant impact on

whether or not Catholics made a donation to the religious in-group.

The analyses show no impact of our five specific religious triggers nor of the general

religion prime on Muslims. There are several possible reasons for this that invite further

14

investigation. First, the literature on experimental primes cautions that if participants already

have a very strong sense of the construct one hopes to manipulate, it may not be possible for the

prime to further activate that, leaving the researchers with statistically insignificant results. The

interviews lend some plausibility to this interpretation, in that Muslim interviewees indicated that

duty to God and community have strong roles in their giving behaviors.

Second, the experiment design had to make some compromises. One was that, in order to

keep the options as similar as possible between the two religions to facilitate comparisons and

equality of treatments, we constrained the list of charity organizations in the experiments. Recall

that the non-student participants were recruited from the Gülen movement. In the experiment,

there was no option for giving to a charity organization specific to the Gülen movement. Some

participants kept their participant payment in order to donate it to their organization later. If a

familiar, Gülen community-affilated organization had been in the list, the number of people who

donated might have been greater. Given the strong identification, voiced in the interviews, that

Gülen members have with their group, it may have been overwhelming and thus washed out

subtleties of the primes. That the Muslim university students gave more than Muslim community

members could also be an indication of this problem. Normally, we expect the reverse (students

to donate less). The university students included in the analyses are Muslims but not necessarily

Gülen members. In fact, if we code our donation measure to include individuals who wrote in

that they would prefer that their donation go to a different organization or that they would keep

the money and donate it later themselves, we find that there are substantially more Muslim

donations, and that community members were more likely to have made a donation than

students, though the individual primes have no statistically significant effect.19

In addition, Catholics were also constrained to donate to a charity that was not specific to

their parish, even though it was an “in-group” charity (Catholic). Rather than write in a

preference, as a number of Muslims did, some Catholics may have shifted their donations to the

out-group charity. It’s possible that Catholics in our sample did not feel the same closeness of

community that the Gülen association members did, or that their own concerns with the on-going

pedophile scandal of the Church led them to donate to the secular out-group. In other words, the

experiment results may have been biased due to the way the donation propensities were captured

in the protocol.

A third consideration is whether the fact of writing an essay was a technique appropriate

for the Irish Catholic samples but not for the Turkish Muslim samples. To test this, we assessed

whether the two populations differed in length of the essays, including number of sentences. The

data rule out this interpretation. There is no statistically significant difference in the number of

sentences written by Catholics and Muslims.

The experiment essays enable us to examine whether specific primes did prompt feelings

or thoughts of generosity and pro-social behavior even though participants were constrained in

their choice of what to give to. As previously described, the subjects provided written responses

to randomly assigned essay primes. These essays were qualitatively coded by research assistants

to examine the impact of the primes on the propensity of subjects to mention charity. Any

sentence that mentioned volunteering, giving of money or offering, or a variety of other

19

However, coding our donation measure in this manner is problematic for two reasons. The first, alluded to earlier,

is that it is impossible to verify whether individuals who mentioned they would take their donations and make

donations later ever had any intention of doing so. Second, because the experimental prompts did not provide an

explicit option for making donations to other religious groups, it is impossible to know whether there were

individuals who simply marked to keep their money who did not in fact then make a donation to a separate group.

15

charitable behaviors was coded as one mention of charity.20

See Appendix ___ for the complete

codebook regarding the qualitative coding of charity.21

We can use these data to see if the primes prompted thoughts of being pro-social

(“charity” in our coding), even if not matched by actual donations. This would lend some albeit

weak support to the related hypotheses. As in the experimental analysis above, the primes are

included in regression analysis as categorical variables. Coefficients for each of the experimental

conditions are interpreted in comparison to the general control condition. The community and

student samples are pooled into each of the Catholic and Muslim populations, and “Sample” is a

dichotomous variable indicating whether the individual was part of the community or student

populations. Ordinary Least Squares regression was used to analyze the content of the essays

given the continuous nature of the number of mentions of charity in the essays.22

[Table 5 about here, see p. 30]

Table 5 demonstrates the total number of charitable mentions made within the separate

Irish Catholic and Turkish Muslim samples. There is remarkable similarity between the Catholic

and Muslim respondents’ propensity to mention charity within their essay responses. The vast

majority of respondents, about 76% of the Catholic respondents and 74% of the Muslim

respondents, did not mention charity. About13% of Catholics and 14% of Muslims mentioned

charity once, 7% of Irish Catholics and 6% of Turkish Muslims mentioned it twice, and close to

5% of Irish and Turkish respondents mentioned charity three or more times.

Table 6 presents the results of the OLS regression assessing the relationship between the

experimental primes and the number of mentions of charity in respondents’ written essay

responses.

[Table 6 about here, see p. 31]

For Irish Catholics, exposure to the Grace and Deservedness primes had statistically

significant impacts on the mentions of charity in the essays. Individuals who received the Grace

prompt were associated with a 0.41 increase in the number of mentions over individuals who

received the Control prompt, while individuals who received the Deservedness prompt were

associated with a 1.87 increase in the number of mentions of charity over the control prompt.

Among Turkish Muslims, the Community, Deservedness, Duty, and Grace primes each had

statistically significant impacts on the number of mentions of charity. Individuals who received

the Community prime were associated with a 0.35 increase in the number of mentions of charity,

while individuals who received the Deservedness prime were associated with a 1.54 increase in

the number of mentions of charity. Individuals receiving the Duty prime were associated with a

0.33 increase in the number of mentions of charity, while individuals who received the Grace

prime were associated with a 0.45 increase in the number of mentions of charity.

The deservingness prime had the strongest effect on both Catholics and Muslims; one

interpretation of the results is that both religions emphasize that those in need are deserving of

help. Another is that the experiment tapped a strong emotional reaction, and that the religion’s

20

Individuals who did not leave a written essay response were coded as having made no mentions of charity. 21

108 Irish and Turkish essays were randomly selected and recoded for checks of inter-coder reliability. These

checks resulted in a Krippendorf’s alpha ratio value of 0.853. 22

The categorical predictor variables, combined with the relatively small sample, precluded the possibility of using

negative binomial regression.

16

connection to it is spurious. A third is that because this prime explicitly asked about help (“Think

about people in need who deserve help,” it of course prompted more frequent mentions of factors

we coded as “charity”.

The results do not tell us whether the participant was thinking of the in- or the out-group

as the target of his/her charity. They enable us to see that Muslims were prompted by the

community expectations, God’s grace, duty to God and deservedness primes to think of charity,

even if they did not go on to donate in a way that was statistically significant. We can only

speculate that under different donation options in the experiment, those primes would have

elicited donations in the expected directions.

The fact that duty to God did not prompt Irish Catholics to write about charity suggests

that that the prime was triggering a different concept that was linked to donating behavior . A

reading of the Irish Catholic “duty to God” essays shows that most community participants

responded that their duty to God is to “love your neighbor” (id # 284).23

Some said it is to keep

the sacraments, or to obey. One interpretation of this is that the duty to God triggers thoughts of

love of others, which may be what triggers the pro-social behavior observed in the Irish Catholic

sample (responding to that prime).

interviews

The second methodological prong of this paper is a set of semi-structured interviews with

religious adherents in their respective parishes and associations. Parishes were chosen also on the

basis of referrals and the process of gaining access varied by diocese.24

Associations were chosen

on the basis of referrals from leading figures of the Gülen movement in Europe.25

Estimates of

mass attendance in the parishes ranged from about 8% of 50,000 in Saint Pierre de Montrouge,

Paris, 15% of 14,000 in Santa Maria alla Fontana, Milan, 25% of 6,000 in Ballygal, Dublin, and

12 % of 2000 parishioners in Saint Esprit, Istanbul. Although we do not have exact numbers for

Turks living in European countries and we do not know what percentage of them are practicing

Muslims, the following data might be helpful to see the size of our associations. Ihsander in

Istanbul has about 350 active members and the association serves about 20000 people. There are

about 450,000 Turks living in Paris and Yunus Emre Cultural Association offers religious

services to about 500-600 people. There are about 4,000 Turks living in Milan and the Alba

23

The first sentence of another essay is “What you do for the poor you do for me [i.e. Jesus] therefore it is a good

thing to help the many charities” (# 335). 24

Access to Santa Maria alla Fontana was by direct referral from a scholar of Catholicism in Italy; access to Our

Lady of Divine Grace Ballygal was by referral from the Moderator of the Curia of the Dublin diocese after we had

been granted permission by the Dublin Archbishop to conduct the research in the diocese; access to Saint Pierre de

Montrouge was by direct appeal to the head priest; access to Basilique-Cathédrale Saint Esprit was by direct appeal

to the head priest. We had explained in our contact letters what kind of parish we hoped to conduct the study in: not

exceptional, not in an extremely wealthy or poor neighborhood, and not in a tourist zone. With only two Latin rite

parishes, there was little choice in Istanbul. Saint Esprit was willing to grant access. Appendix B has more

information on the parishes and on the Muslim associations. 25

Şerafettin Pektaş, the President of Intercultural Dialogue Foundation in Brussels, put us in contact with the

associations in Paris, Milan and Dublin. We chose the Yunus Emre Cultural Association in Paris since it was the

largest among the six Gülen community centers in Paris. The association is located in the northwest suburbs of

Paris. For Milan and Dublin we had only one association available in each city. We conducted our fieldwork at the

Alba Intercultural Association in Milan and at the Turkish Irish Educational and Cultural Society in Dublin. The

cultural centers in Milan and Paris are both located at the city center. Muharrem Atlığ of Journalists and Writers

Foundation, a leading Gulen affiliated civil society organization, put us in contact with Ihsander in Istanbul.

Ihsander is located in the northwestern suburbs of Istanbul.

17

Intercultural Association serves about 200 people who are affiliated with the association. There

are about 5,000 Turks living in Ireland, most of whom live in Dublin. The Turkish-Irish

Educational and Cultural Society serves to the Turkish community with about 100-125 people

affiliated with the association. The interviewees comprise convenience samples drawn from

referrals from religious officials, referrals from volunteers to whom a religious official referred

us, and some ad hoc button-holing of religious service participants. We had explained the project

and the range of individuals we sought to interview to those who referred us but were unable to

control whether the individuals were from some ideal range. Many of our interviewees are more

involved in their religious organizations and more frequent practitioners of the formal rites of the

religions, but not all. To facilitate participation, the interviews were conducted at locations of the

interviewees’ choosing. This was typically a room in a building owned by the parish or

association, in the interviewee’s home, or in a café or restaurant. Some interviews were held at

the interviewee’s place of business.

The semi-structured interviews lasted between 30 minutes and two hours, and included

questions about the interviewee’s connections to the parish or association, forms of participation

in parish and other volunteer activities, what the teachings of the religion are on generosity, if

they remembered sermons that spoke to a theme of helping others or being generous, what kinds

of appeals motivated them to give funds or time, and, directly, why they were generous, whether

they thought they had a responsibility to God to help others, what happened if they weren’t

helpful, what kinds of responsibility, if any, individuals, the religious organization and the state

had to help those in need, what was inspiring them or motivating them when they were helping

others, giving of themselves. We invited them to comment on anything we should have asked

them about but didn’t. With the exception of some Catholics in Istanbul, each interview was

conducted in the mother tongue of the interviewee.26

Participants were asked to fill out a survey

asking demographic information and assessing religiosity.

We also conducted interviews with leaders of each religious organization (priest, bishop,

imam, association leader) to learn how they think and reason about generosity and to enable us to

document the broader context and considerations of the giving and voluntarism in the religious

organizations. We gathered information on funding structures, as well as volunteer programs,

and engaged in participant observations of weekly religious services and charitable activities.

Catholics and Duty to God:

What we find for Catholics, and what we find in some of the experiment essays, is that when

Catholics are asked if they have a duty to God to help others, they say “no”, and instead turn the

duty to a duty to help others and a choice to live as God tells them to live: that is, following

Jesus’ example. A number of the experiment essays by Catholics show that they literally made

that move when asked to think about their duty to God. These facts together may explain why

“duty to God” could have a strong effect in the experiments but not be a phrase or concept

Catholics recognize as applying to them in regard to helping others. The faith does not rely on a

sense of duty or obligation to God. As a male Italian interviewee in his 30s stated (# 13), “Jesus

didn’t say ‘you have to do this.’ Jesus said ‘this is the path, you choose.’” An Irish parishioner

echoes this view: “God gives me a choice. God prompts me and he allows me to say yes or no

26

Some Catholics in the Istanbul parish were immigrants from countries of which the interviewer did not speak the

home country language. In those cases, the interviews were conducted in Turkish or English, depending on the

preference of the interviewee.

18

and I know if I say yes he’s going to be there with me all the time. There’s nothing he won’t help

me with” (#79). The religion relies on love, on love of Jesus, on love of the other. It needs to be

stated that priests and many interviewees see this love as a source of charity towards others, but

that they indicate (expressly or implicitly) that is not the point or goal of such love. The love is

an end in itself, that also and inevitably leads to generous behavior by believers. One interviewee

(#11) cited the Gospel (Matthew 25:40), “Inasmuch as ye have done it unto one of the least of

these my brethren, ye have done it unto me.” An Irish parishioner stressed that helping and

giving is “not a duty, there have been opportunities. I was drifting a bit, the door opened” (# 28).

Another responded, “you’re not morally bound, I suppose, to give… It’s part of Catholic

teaching to be charitable, part of your religion and belief that you must help the deprived.” He

used the phrase “must help” but he rejected the idea that helping others was a duty or obligation

to God (# 82). It was a general “you should” and not one mandated by God. Because Jesus’ and

God’s love does not have to be earned, earning it through giving to others (or to God via

donations to the church) is irrelevant. It doesn’t make sense in the religion. A parishioner from

Istanbul (# 13) said “God loves those who followed what Jesus did” but she did not put this as a

requirement to gain God’s approval. Parishioners who linked their giving with specific scriptures

often noted, as one flatly stated, “there are only two commandments that matter: love God and

love your neighbor” (# 80-1). For our Catholic interviewees, that was the source of all

generosity, and hence collective goods provision.

Muslims and Duty to God

Islam emphasizes obedience to God’s will. If God’s will is that the individual be

generous, then religiosity affects generosity by raising Muslims’ sense of an obligation to God.

[…] It is “for the love of Allah” that one gives to others (Quran 2:177). This belief is a

significant factor for Muslims’ pro-social, cooperative and generous behavior. Though not

enforced by humans, it acts as a strong incentive and sanctioning mechanism. The vast majority

of Muslim interviewees indicated that they gave for the sake of God. Almost all the Gülen

association interviewees understood their giving as due to their religious conviction that giving is

a duty to God. Most respondents had the belief that what they own on the world is temporary;

resources are given to them by God to test the believers. To them, how they use these temporary

resources here will have a big impact on their situation in the afterlife. They think that when they

give, they give from God’s property. (e.g. Istanbul 2, Istanbul 20, Paris 7). A quotation from an

interviewee from Istanbul (#7) illustrates this point: “We do not see money and our belongings

as our real property. We see them as God’s deposit on us. We can fulfill our responsibility to that

deposit if we use it properly. Giving for the sake of God is one way of fulfilling that

responsibility.” Some respondents quoted the following verse: “Behold, God has bought of the

believers their lives and their possessions, promising them paradise in return” (Quran 9/111) (e.g.

Paris 25). A businessman echoes this point with the following words: “When we give to others,

we actually invest in ourselves. At first sight, it looks like we did something for someone else.

But in reality, we discharge a big responsibility by transferring God’s property to those in need”

(Istanbul 18).

A significant percentage of interviewees argued that if they do not give, they lose the

resource one way or another anyway. For example, one of the interviewees said that whenever he

refuses to help to his association, he is caught by a traffic policeman and has to pay a fine that is

the amount of money he otherwise should have given his association (Paris 3). The same view

was echoed by an Istanbul Gülen association member (#13), “If you do not give to the places that

19

you are required to give, that [amount] will be taken from you eventually. You will have a

problem and lose it anyway.”

Catholics and Community

The interviews provide insight into the failure of community expectations to lead

Catholics to be more generous. Catholicism operates not with group imposed expectations by

encouraging and providing an outlet for individuals’ pro-social orientations. Individuals who

volunteered often voiced that they liked to be with people, to be focused on something, to be

collaborating. Community is an important means of fostering their provision of public goods.

Yet the means depend on attitudes towards the (religious) group, not sanctions or incentives.

People are attached to the parishes they were interviewed at, they feel loyal to them, and they

like working in the groups they are in or have volunteered with before.27

This attitude of

attachment (“affect”) may have influenced their willingness to be involved and contribute,

without them being aware of it. The head of the St. Pierre de Montrouge parish Economic

Council explained that the diocese of Paris imposes a tax of about 16% on the parish in order to

redistribute its wealth, and to use its wealth for some of the diocese’s operating expenses. He

said that a number of parishioners have come up to him and said “I hope that what I give doesn’t

go to the diocese.”28

Some Irish parishioners expressed concern over the clustering of parishes

(as did the priest at Ballygall), anticipating that the community spirit of the parish would be

diminished, and that people might not be as willing to volunteer or give if their efforts went to a

bigger entity less known to them. One parishioner there noted that “the first collection [at mass]

is for the priests (and even some of that goes into the archbishop’s house which people don’t

know and I think they should). I have a little bit of difficulty with the two baskets, especially the

second one, which is for share. It’s for building churches in Dublin. There’s no churches that

have been built. So I don’t know where the money goes. I really don’t know where the money

goes. It goes into the Archbishop’s House but I don’t know what they do with it” (# 79). Many

parishioners expressed satisfaction with their experiences of working with other parishioners in

the various organizations they volunteered in, and positive affect towards the priest, sister (nun)

or friend who invited them to get involved. Some saw volunteering within the parish or with a

Catholic charity as furthering the work of Jesus or of the Church in the world. This was also very

evident among the congregants of the St. Esprit church in Istanbul. While the older and

Levantine volunteers mostly referred to their loyalty to the church (Istanbul Catholic 5, 6, 20),

the younger and immigrant members mostly addressed the friendship within the church (Istanbul

Catholic 10, 23, 24). In the words of a Filipino woman who migrated to Turkey in the 1990s,

“when I first came to Turkey, a friend brought me to this church. I felt at home and since then I

contribute either financially or volunteering in the church” (Istanbul 3). For most of the

immigrants the local church is seen as a place where their problems are addressed within small

friendship groups. But not contributing or volunteering had no negative community-level

consequences.

While the religious community might facilitate Catholics being able to volunteer and give

in certain ways, it would be a stretch to say it directly creates, as an incentive, one of the key

factors interviewees mentioned when asked why they give or volunteer. Most said they got back

more than they gave, that they thought it was more rewarding for them than for the people they

27

This is not exclusive to Catholicism or religious groups. There are a number of findings converging in the social

sciences on the positive role attitudes towards the group, towards the task and affect/emotion towards the group play

in motivating pro-social, cooperative behavior (Tyler 2011). 28

Yves Jouen, Feb. 24, 2011

20

helped, that they learned from those they helped. “I get repaid abundantly” (Italian woman in late

50s, # 11). A number of interviewees, especially the Irish, emphasized they “feel fortunate”, and

by helping are “giving thanks for what you got” and that because of this they want to “help the

people who aren’t” healthy or otherwise “ok” (# 30). A retired Frenchwoman said she feels she’s

been “spoiled by life” and got involved because she wanted to “give back” some of what she had

received (# 42/43). An Irish woman who for a long time had the sense she’s been blessed, and

was fortunate, stated that there are people “for whom life is a lot harder and if I could help

people, I’d like to” (#28). A retired woman in Istanbul noted her happiness when she saw “the

needs of the poor immigrants are met” within their volunteer group (Istanbul Catholic 16). In

addition, many Italian and French interviewees mentioned that the interaction between “giver”

and “recipient” itself was critical. For them it was a fundamental point of generosity—not just to

give but to interact with the recipient.29

It is possible that these orientations account for the

tendency of Catholics to respond to the deservingness prime in the experiments even though they

did not respond to the community prime.

Those who volunteered in activities that sustained the religious life of the church, such as

the children’s ministry, reading the liturgy, being on the baptism or funeral team, often noted that

doing so enabled them to attain a greater understanding of their faith, and noted that was

important and rewarding to them (e.g. # 28, 50s Irish woman). Some also felt they had a

responsibility toward the community: it needed the aid of each to run (# 30, 30s male):

“everyone should play a part, not just one guy doing five jobs”. Similar responses were given by

the Catholic minority in Istanbul. The participants of a volunteer group (the Legion of Mary)

within the church indicated how they deepen their friendship within the group and convert it into

good things (Istanbul Catholic 3, 10, 25). The women in this group (there was only one male

member) were mostly immigrants and they were both contributing to the activities organized by

the church and their newcomer immigrant friends. Yet there are no mechanisms to sanction (or

identify) the free-riders.

It was clear from the interviews that Catholics both supported their community

financially and through volunteering, and supported those who were not part of the community,

either by direct donation to a non-Catholic charity or through a Catholic-affiliated charity (or by

helping their neighbor, regardless). These findings contrast with the experiment results, that

showed a distinct Catholic tendency to donate to the out-group (though not with the community

expectations prime). The interviews and the experiment essays again provide some insight on

these unexpected findings. Our field research coincided with a horrific pedophile scandal by the

Catholic Church in Ireland. Interviewees, including the priests and the moderator of the Curia,

brought up the scandal. Interviewees and experiment participants who mentioned it (the latter in

their essays) said they were ashamed of their Church. They had good feelings about their own

parishes, but not the institutional church. Because the experiment prime was for a donation to a

29 In most cases, at the time of our study, the structures necessary for public goods provision were already in place

in the parishes and dioceses. We are not able to address, in this study, the conditions under which Catholicism, as an

organized religion, provides the basic structures. Clearly, the cost of being a volunteer is lower if the structures are

already in place. Yet we do have a few instances, and these cohere with the general tenor of our findings: either a

priest decides to launch a program, and asks for volunteers, or one or more parishioners suggest a project to the

priest, and he agrees, gives them some resources from general parish funds, and then they work on developing it,

and asking for more volunteers.

21

Church organization, and not to the local parish, participants may have steered their donations

away from the in-group charity.

The interviews also revealed that Catholicism relies extensively on the faith, and/or

upbringing or other life situations, to give Catholics a disposition to be generous, rather than

mostly on a set of incentives and sanctions based on material factors. All the priests interviewed

spoke of an emphasis on faith and on having parishioners understand their faith as being the

means of promoting generosity. Many parishioners spoke of their family background and

upbringing, or schooling (granted, it was in Catholic run-schools—older Irish generations

remembered being told they had to help “babies in Africa” even if, as schoolchildren, they didn’t

know why) as being the source of their generous dispositions, their sense that they wanted to

help others in need. While noting her parents were very devout Catholics, an older Irishwoman

notes: “I was brought up in a very caring and giving family … some families give and some

don’t. A huge amount of it goes down to my mom and dad, and my aunts and uncles” (# 79).” A

French interviewee, a devout Catholic raised in the Church and who frequently attended mass

and led some volunteer groups in the parish, credited his participation in Boy Scouts as giving

him his sense of wanting to help others (# 55).

Muslims and Community

As exemplified by the Gülen adherents, religious community ties are important to

Muslims. The importance is both in belonging -- Gülen members see themselves as part of the

global initiatives of the movement-- and in the expectations they feel within the community. We

turn first to the sense of belonging. For example, even though the Istanbul community assists

non-Gülen charities, they , like Catholic parishes with their organizations, do this through the

movement’s own charitable agency (Kimse Yok Mu Solidarity and Aid Association, or KYM).

They also have a feeling of ownership in many of the institutions that the movement has opened

elsewhere. The most frequently cited global endeavors of the movement were schools

established in poorer regions of the world (especially Africa), the intercultural activities of the

organization, and the sacrifices of the younger recruits of the movement (especially graduates of

prestigious universities in Turkey who goes to very poor countries of the world as teachers).

Most interviewees felt their relationship with the movement special, and their membership to the

movement emotionally influenced them. The contributions they make feel them the builders of

many global institutions of the movement. When they talked about charity, the movement-led

educational institutions were the first things that came to the interviewees’ minds. Even when

they think of poverty, they refer to the movement’s charitable organization, KYM, and note how

it works to relieve poverty. This was especially evident in Istanbul and Paris, probably because

the contributions in these cities are used to support global projects of the movement. The

contributions in Italy and Ireland hardly support within-country costs of the movement due to the

small number of supporters living in these countries.

The interviews revealed that two things strengthen Muslims’ sense of community. First,

the movement always informs its followers about their activities. Most subjects (almost 80 %)

stated that to see the movement’s local and global activities motivated them the most. Most

subjects in Istanbul and Paris (where the movement is large and have relatively richer followers)

stated that they had foreign trips to see the activities of the movement abroad. Being informed

increases the bonds of the followers to the movement (e.g. Istanbul 21, 29; Paris 1, 5, 7). A

respondent from Paris (# 7) mentioned how his trip to Pakistan organized by the association to

help the victims of Pakistani flood deepened his relationship with the movement. Second, the

lack of strong hierarchical structures increases group identity and awareness. Since each local

22

setting is autonomous and self-funding. Adherents make the investment decisions in their own

local setting and they try to finance it with the local sources. They have a feeling of ownership

on what is being done in a certain place. This, in the end, strengthens their attachment to the

association.

In addition to this, members also have close relationships among one another. These

friendships might also have a role in their sense of community and giving. Most of the

interviewees indicated that their involvement in the associations created a social environment for

them. Most interviewees see other members of the association at least once weekly and they

work on various projects together, thus developing social bonds. A furniture storeowner from

Istanbul (# 17) mentions friendship and giving equally: “Being involved in this movement made

giving part of our personality. It is because you become part of a new social environment. Here,

we compete for good things […] The place that I like the most is my village. I love to live there.

Sometimes, I think about going back to my village and living there happily. I have money and do

not need to work more to earn my living. However, when I go to my village just for a visit,

within two weeks I miss my friends. Also, I keep working here to give more. I cannot stop

working in the face of the needs of so many people.”

The interviews reveal that community expectations are at work in prompting Muslims’

club and public goods contributions. In contrast to the Catholic parishes, most of the giving in

the Muslim associations is public. The annual pledges, that constitute the major source of income

for the associations, are entirely voluntary. Each contributor stands and announces publicly at the

fundraising even what his or her pledge is, so that all the attendees in the room know how much

others say they will contribute. This gives Islam (or at least the Gülen movement)30

a relatively

powerful peer pressure sanctioning (and incentivizing) mechanism-one has an incentive to gain

community approval by giving, and one is concerned about the sanction of community

disapproval if one doesn’t give. Statements from two interviewees in Istanbul (# 14 and 17)

illustrate the point. One, a middle-aged man (# 14), when explaining his giving during the

pledges explained that he compares his giving with others: “I see other people who have

circumstances similar to mine giving more than I do. This encourages me to give more.” An

association member in Istanbul (17) stated, “We compete for good things,” and emphasized the

social dimension of giving. He used a Turkish proverb when explaining the impact of peers on

giving: “Üzüm üzüme baka baka kararır”: “Grapes will darken by looking at each other.” Other

interviewees mentioned the hadith, (a saying of the Prophet Mohammad): “Compete for the

good.” Yet it was not evident that individuals who did not give much or who gave nothing at all

were given a cold shoulder at community functions or otherwise looked down upon. When

talking about the pledges, the respondents referred to them as “incentives,” rather than

“sanctions,” (Istanbul 18) and noted the practice’s origins in Islamic history origins (Paris 7). In

the words of a respondent from Istanbul, (# 18) “Organized programs by the movement of course

influence our giving. If we were alone, we would not give this much. We should remind each

other of our religious duties to each other. This is what the movement does in a systematic way.”

The collection of the pledges is monitored through weekly meetings throughout the year. A

religious association official explained that when followers pledge “they also state how the

30

Kilinc observed that other Islamic associations/mosque also have a similar practice. He observed that the mosques

in Tempe, AZ, East Lansing, MI and Omaha, NE pursue a similar strategy in the month of Ramadan. When asked

about the origins of these pledges, we were told that this practice originated in the time of Prophet Mohammad.

When there was a need, the public fundraising was organized and the companions of Mohammad competed one

another for the good (Istanbul 24, Paris 30).

23

payments will be made. It may be an advance full payment or paid monthly or paid in the form

of a few installments throughout the year. […] We kindly remind the contributors if it is not paid

on the promised time. If they are not able to pay, of course, there is nothing to do” (Paris 24).

The statements of a respondent from Istanbul (# 22) show that “duty to God” comes into play in

fulfilling the pledges as well: “When we pledge an amount, we feel that we promise God to give

that amount. When we make the payment, we feel we are fulfilling our promise to God.” It is

through the community that the link to duty to God is made.

As with the Catholic parishes, the tangible benefits provided by the Islamic association

are not exclusionary. They are basically public goods, not club goods.31

In Istanbul and Paris, the

association has private schools. These are tuition-based schools, and those sending their children

pay the same tuition whether they contribute and belong to the movement or not. For example,

the school in Paris was new. They were looking for students to keep the school financially strong

so had no need to exclude any on any grounds. The concern was not on children being excluded

from attending the schools, but on what would happen to the children if the schools didn’t exist.

Interviewees sometimes voiced a concern that if they did not donate funds, their children in the

words of one, “would not have proper facilities and would be in danger of losing their identity”

(Interviewee # Paris 28). The concern was about the association not having enough resources to

establish and maintain the institutions, not a fear of being excluded if one did not contribute. The

Gülen movements not only allow others to attend the activities of the association and send their

children to the school, they also expend considerable effort to have as many people attending,

including children at the weekend school, as possible. As with the Catholic parishes (including

on religious education where not provided by the public schools), there is no difference between

contributors and non-contributors in benefiting from the activities of the associations.

After reviewing the interview findings in light of the experiments, what should be noted

is that Catholicism and Islam are each religions that emphasize the community and the beliefs—

the community is there to sustain the believer, many of the religious rituals are to be done jointly

with other believers, the community is a spiritual community, the religion inheres in a religious

community. The social and spiritual experiences of the faithful are tightly linked for most, and

not easily separated into distinct categories.

Conclusion

We have suggested and demonstrated that religious beliefs and community structures

affect the club and public goods provision, and more generally, pro-social tendencies, of

individuals through specific mechanisms. While we are in agreement with rationalist theories

that individuals make cost/benefit decisions that determine their level of engagement with regard

to the production of both club and public goods, we also accept the challenges from a

multidisciplinary body of literature suggesting that individuals have pro-social tendencies, and

we suggest that religions, through their beliefs and communities, can trigger these in ways that

are systematic and observable. Rather than viewing these pro-social behaviors as exogenously or

31

For the Muslim schools in Paris and Istanbul (there was no school in Milan or Dublin), one does not have to be

Muslim to attend. The school in Paris had non-Muslim students as well. The Gülen association was working to sign

a contract with the French state to get some financial support. However for the weekend school of the associations,

this is not the case. The weekend schools in all four cities were solely for religious instruction. There was not a set

rule that only Muslims could attend. In practice, it is highly unlikely that a Christian would send their children to

such a weekend school. However, in Milan, children sometimes bring their Italian friends to the cultural center

(Milan 21).

24

randomly distributed across the population, we contend that religions can play a role in

systematically influencing the form of other-regarding preferences, and that religions could

produce varying patterns in how other-regarding preferences are expressed. We find that there

are differences in the impact of similar belief concepts on Catholics and Muslims. But we also

find that religious beliefs seem to override expected socio-psychological proclivities to engage in

the production of club and not public goods. Instead, both religions produce public goods

through activation of religious beliefs and experiences such as duty to God, God’s grace, and

deservingness. While the experiments showed that there were effects for these factors on

Catholics, and none on Muslims, the interviews show that duty to God is a strong factor in

Muslims’ pro-social behavior, and that it is not understood as a duty to God by Catholics. We

have some tentative hypotheses about the reasons for these discrepancies, having to do with the

specifics of the experiments and the specific populations they were conducted on. The interviews

also showed that community affect is a strong factor for adherents of both religions.

The two methodologies revealed some differences in outcomes from the same variable.

We have discussed possible reasons for this. In the experiments, the various religion primes did

not affect Muslims in a statistically significant way at all. That said, statistical analysis of the

essays indicated that several primes, duty to God, community expectations, God’s grace, and

deservingness, tended to generate thoughts of charity. The same analysis of the Catholic essays

showed that the latter three primes prompted thoughts of charity, and that duty to God did not.

This is consistent with Catholic’s understandings and theology about the sources of generosity

that we argue underlies public goods provision. In the interviews, it was striking how the vast

majority of Muslim interviewees described their motivations for helping and giving funds as

stemming from a duty to God and expectations of the community. They also, like Catholic,

indicated the importance of community affect: positive feelings towards the group and pleasure

in being active in it. Finally, the interviews showed that both religious communities are weak on

external monitoring and sanctioning devices. The Gülen association’s is limited to publicly

announced annual pledges and to some sense of being compared to other members in level of

donations and in extent of volunteering; the Catholic parishes had no such public statements or

sense of competition among parishioners. Muslims vocalized a sense of God monitoring and

sanctioning them; Catholics did not, even though the experiment prime of duty to God induced

generosity (while the essays they wrote showed that they turned duty to God into duty to

neighbor).

We see in the results evidence that the rational choice approach to public goods provision

needs to take into account the inherent pro-social tendencies of individuals, and to pay more

attention to how belief systems and the communities organized around those beliefs can

systematically affect collective action and cooperative, helping behaviors. By focusing on two

religions that are not strict or extreme, in contrast to the literature that has developed about club

and public goods provision by religions, we have been able to see that monitoring and

sanctioning, and material incentives, do not have to be the primary means by which religions, or

perhaps other organizations, get their adherents to be cooperative and helpful, indeed generous.

Instead, beliefs and positive community affect can be important factors in engaging and

channeling the inherent pro-social tendencies of individuals. While there has been a drive in

some of the scholarly literature to argue that what drives the pro-social behavior of the religious

is their involvement in a [strict] religious community, and not their religious beliefs, we find that

our subjects do not make such distinctions. As Durkheim stated, and others have elaborated on

25

(Lincoln 2003), religious beliefs are closely linked to the communities—the one inheres in the

other.

Religions do not have an oligopoly on explaining the non-governmental production of

public goods or cooperation between individuals. They are one set of likely producers. However,

because they are nearly ubiquitous in the organization of human life, understanding the

mechanisms by which organized religions potentially contribute to the production of public

goods provides us with important insights into a significant part of human behavior.

26

Table 1a: Proportion of Those Making a Donation

Donation Type N Freq. %

Any Donation 809 230 28.43

Ingroup Donation 809 121 14.96

Outgroup Donation 809 172 21.26

Table 1b: Proportion of Those Making a Donation – Catholic and Muslim Samples

Catholic Sample Muslim Sample

Donation Type N Freq. % N Freq. %

Any Donation 337 132 39.17 352 69 19.77

Ingroup Donation 337 56 16.62 352 59 16.76

Outgroup Donation 337 106 31.45 352 38 10.8

27

Table 2: Logit Analysis: Whether Any Donation was Made

Catholic Sample Muslim Sample

Prime Odds Ratio Standard

Error Odds Ratio

Standard

Error

Sample 0.30*** 0.08 2.97*** 0.82

Community 1.49 0.67 0.87 0.45

Self/Other 1.19 0.54 0.90 0.47

Deservedness 1.98 0.86 1.29 0.64

Duty 2.62** 1.19 0.64 0.35

Grace 2.27* 0.97 1.15 0.57

Gen. Religion 1.48 0.68 1.22 0.63

coefficients reported as odds ratios

Catholic N = 337; Muslim N = 352

*p < 0.10; ** p < 0.05; *** p < 0.01

28

Table 3: Logit Analysis: Whether InGroup Donations Were Made

Catholic Sample Muslim Sample

Prime Odds Ratio Standard

Error Odds Ratio

Standard

Error

Sample 0.34*** 0.13 2.58*** 0.76

Community 0.85 0.48 1.00 0.55

Self/Other 0.75 0.44 0.88 0.50

Deservedness 1.11 0.60 1.34 0.71

Duty 1.48 0.80 0.57 0.35

Grace 0.76 0.43 1.36 0.71

Gen. Religion 1.35 0.73 1.62 0.87

coefficients reported as odds ratios

Catholic N = 337; Muslim N = 352

*p < 0.10; ** p < 0.05; *** p < 0.01

29

Table 4: Logit Analysis: Whether OutGroup Donations Were Made

Catholic Sample Muslim Sample

Prime Odds Ratio Standard

Error Odds Ratio

Standard

Error

Sample 0.39*** 0.11 5.85*** 2.28

Community 1.83 0.90 0.97 0.69

Self/Other 1.94 0.96 0.79 0.57

Deservedness 2.72** 1.29 1.48 0.96

Duty 2.88** 1.42 0.75 0.54

Grace 3.18** 1.49 1.32 0.85

Gen. Religion 1.51 0.78 1.48 0.99

coefficients reported as odds ratios

Catholic N = 337; Muslim N = 352

*p < 0.10; ** p < 0.05; *** p < 0.01

30

Table 5: Mentions of Charity by Sample

Irish Catholic Turkish Muslim

Mentions of

Charity Freq. % Freq. %

0 254 75.82 260 74.07

1 44 13.13 50 14.25

2 22 6.57 21 5.98

3 8 2.39 12 3.42

4 2 0.6 4 1.14

5 3 0.9 4 1.14

6 1 0.3 0 0

7 1 0.3 0 0

31

Table 6: Primes and Mentions of Charity

VARIABLES

Irish Catholic

Sample

Turkish Muslim

Sample

Sample -0.0176 0.0666

(0.0913) (0.0944)

Community 0.260 0.352**

(0.160) (0.168)

Self/Other 0.127 0.101

(0.159) (0.169)

Deservedness 1.865*** 1.539***

(0.155) (0.169)

Duty 0.193 0.332**

(0.165) (0.168)

Grace 0.407*** 0.450***

(0.154) (0.166)

Gen. Religion 0.0670 0.210

(0.161) (0.175)

Constant 0.00633 0.0154

(0.115) (0.123)

Observations 335 351

R-squared 0.394 0.243

Standard errors in parentheses

*** p<0.01, ** p<0.05, * p<0.1

32

PROFILE OF THE INTERVIEWEES for the semi-structured interviews

Table 7: Basic Demographics of Muslim Interviewees

Muslims in

Istanbul

Muslims in

Paris

Muslims in

Milan

Muslims in

Dublin

Total Number 31 31 32 30

Female 12 10 8 6

Male 19 21 24 24

Age 18-24: 1

25-36: 20

37-55: 9

56-70: 1

71-: 0

18-24: 1

25-36: 15

37-55: 14

56-70: 1

71-: 0

18-24: 4

25-36: 18

37-55: 7

56-70: 0

71-: 0

18-24: 3

25-36: 18

37-55: 7

56-70: 0

71-: 0

Socio-economic

Status*

1: 0

2: 14

3: 10

4: 1

5: 2

No Ans: 4

1: 3

2: 10

3: 8

4: 5

5: 0

No Ans: 5

1: 0

2: 13

3: 8

4: 7

5: 0

No Ans: 4

1: 1

2: 6

3: 3

4: 9

5: 1

No Ans: 10

Table 8: Basic Demographics of Catholic Interviewees

Catholics in

Istanbul

Catholics in

Paris

Catholics in

Milan

Catholics in

Dublin

Total Number 28 21 21 24

Female 9 13 12 16

Male 19 8 9 8

Age 18-24: 0

25-36: 9

37-55: 10

56-70: 8

71-: 1

18-24: 1

25-36: 1

37-55: 6

56-70: 6

71-: 1

18-24: 1

25-36: 3

37-55: 9

56-70: 7

71-: 1

18-24: 0

25-36: 1

37-55: 9

56-70: 10

71-: 4

Socio-economic

Status*

1: 0

2: 11

3: 10

4: 7

5: 0

No Ans: 0

1: 3

2: 9

3: 3

4: 1

5: 0

No Ans: 5

1: 0

2: 6

3: 9

4: 0

5: 0

No Ans: 7

1: 0

2: 22

3: 1

4: 0

5: 0

No Ans:1

* self-reported socio-economic status

1: Lower class

2: Lower middle class

3: Middle class

4: Upper middle class

5: Upper class

33

Appendix A: Demographic statistics on experiment participations in Ireland and Turkey

In Ireland, 62.5% of the population was female, while 37.5% of the Sample was male - that

skewed a bit female. In Turkey, the sample was more balanced, with 52.22% female and

47.78% male.

In Ireland, age breaks down as follows:

18-24: 45.74%

25-36: 5.6%

37-55: 19.95%

56-70: 21.9%

71+: 6.81%

In Turkey, age breaks down as follows:

18-24: 46.65%

25-36: 31.96%

37-55: 19.85%

56-70: 1.55%

71 + : 0.77%

Social economic status, Ireland:

Lower Class: 3.9%

Lower Middle Class: 16.83%

Middle Class: 59.76%

Upper Middle Class: 18.54%

Upper Class: 0.98%

Social economic status, Turkey:

Lower Class: 2.05%

Lower Middle Class: 7.6%

Middle Class: 61.11%

Upper Middle Class: 19.59%

Upper Class: 9.65%

Marital Status, Ireland:

Single: 58.29%

Married: 34.88%

Other: 6.83%

Marital Status, Turkey:

Single: 49.21%

Married: 48.94%

Other: 1.85%

34

Appendix B: The Catholic parishes and Islamic associations

CATHOLIC PARISHES

Santa Maria alla Fontana, Milan, Italy.

This is the parish in which Silvio Berlusconi was baptized (he was not a billionaire at the time).

The neighborhood, hemmed in by the Zara, Isola and Corso Como areas outside the tourist zones

of Milan, is modest, comfortable, and a vibrant mix of Italians and immigrants. The parish has

14,000 parishioners, 2000 attend mass weekly. It has four priests, including one who is retired

but who helps with functions. The volunteer groups include: Saint Vincent de Paul (SVdP, San

Vincento di Paolo), l’Oasi (retired peoples’ group), the Oratorio (youth center), the Pastoral

Council, and catechism school. Others volunteer (and there is overlap between groups) in the

parish office, and with the liturgy, and with various celebrations that take place at the church

(annual saint day, Christmas sale, etc). In the parish but not run by it is a Food Bank, staffed by

many volunteers who attend the church. The church is largely sustained by funds collecting

during offerings and by pledges. For some restoration work (the basic work), the church has

gotten funds from the Italian state, due to the church’s status as an historic site. The church also

has special collections during the major religious holidays (particularly Christmas and Easter)

that go to particular causes, such as the poor in the parish, or for a school run by missionaries in

Africa. Parishioners are notified in various ways of funding requests and needs: mostly the

monthly bulletin that can be picked up at the church, and the posting of the monthly budget

revenues and expenses on a bulletin board at the main entrance to the church.

St Pierre de Montrouge, Paris, France.

Located in the non-touristy 14th

arrondissment, it has about twelve paid staff and between 500-

600 volunteers just to run the church office and its main organizations. There are 50,000 people

in the parish, about 4,000 attend mass weekly (with about 2100 coming on weekends, excluding

the major holidays). The parish is largely middle-class, upper middle-class, with an immigrant

community as well. It has six priests and one deacon. There are lots of volunteer groups, many

very active. As in Santa Maria alla Fontana, many parishioners volunteer with other Catholic

charities in the parish, that are not run by the parish (Secours Catholique, Saint Vincent de Paul

or SVdP, Comité Catholique contre le Faim et pour le Développment or CCFD). They are given

some support by the parish, depending on the preferences of the priest and advice of the parish

Economic Council. Some 80% of its expenses have to be met by tithing and other collections.32

Our Mother of Divine Grace Ballygall parish, Dublin Ireland.

Ballygall is outside the city center, within Dublin (Dublin 11), in a hilly middleclass

neighborhood. It has about 6,200 parishioners (90% Catholic), about 25% (1,300) of the

Catholics attend mass weekly. In contrast to those in Milan, Paris and Istanbul, Dublin parishes

are known by the name of the parish, not by the name of the local parish church. Two priests, a

sister, one sacristan. Parish secretary (paid). Lots of volunteers in a variety of organizations, and

the laity is quite involved in religious services (the liturgy, the children’s mass), in baptism

teams, in funeral team. They are beginning to get trained to do healing services in response to the

paucity of priests. As are all parishes in the Dublin diocese, Ballygall is being clustered with four

others, with the aim of sharing resources. Again, this is due largely to the dearth of priests. The

32

Curé letter & interview with Economic Council head.

35

parishes have had concerns about getting volunteers for activities, and there are worries about

whether the next generations will be as engaged. The Ballygall parish was not one in which any

of the child abuse/pedophile cases were in, but many of the interviewees and the priest

voluntarily brought up the scandal and discussed how hard it has been on the victims, on them,

and how it has led many Catholics to quit coming to mass. The Moderator of the Curia also

brought it up.33

The interviewees who broached the subject always distinguished between the

local parish church, toward which they retained favorable attitudes, and the institutional (or

diocesan) Church, toward which they were critical.34

Basilique-Cathédrale Saint Esprit, Istanbul, Turkey. There are about 2000 Catholics living in

Istanbul. The Basilique Cathedrale Saint Esprit has about 200-250 congregants. The church has

three priests. We were told that a fourth priest had just retired before we arrived. The Head Vicar

also resides in this Church as this is the center of the Vicarate. The church attendees are from

three different groups of people: Levantines (the French and Italian immigrants inherited from

the Ottoman Empire); immigrants from the Philippines, most of whom are nannies; and

immigrants from African countries, especially from Congo. The church also receives many

Catholic tourists because it is an historically important church, and it provides religious services

to the tourists as well. In the summers, there is a service everyday at 6:00 p.m. in the church. The

immigrants, some of whom are in Turkey illegally, live under difficult conditions. Most of the

personal charities go to these immigrants’ needs.

MUSLIM ASSOCIATIONS

Ihsander, Istanbul, Turkey. Ihsander is a benevolent businessman’s association, affiliated with

the Gülen movement, offering religious services to its followers unofficially. It also runs several

charitable activities. Because organizing around religion is legally not allowed in Turkey,

Ihsander cannot have official status as a religious association or organization. The association

has about 350 active members; however, it offers religious and charitable services to thousands

of other people. When the interviews were being conducted, the most pressing agenda was

construction of a dormitory that would house more than one thousand high school students. The

members were trying to gather donations from the people around them to facilitate the

construction.

Yunus Emre Cultural Association, Paris, France. The Turkish population in France is about

600,000. Almost three quarters of them live in Paris. Gulen community has 6 associations in

Paris. The Yunus Emre Cultural Association offers religious and cultural services to its

members. It serves about 500-600 people. Some of the activities that the association organizes

are: weekly religious education courses, weekend school for the Turkish schoolchildren in which

they are taught about Islam and Turkish culture. The association also organizes charitable

activities to help people around the world. When we were there, they organized a program in

which they collected help for Pakistani flood victims. The organization is also affiliated with a

private Turkish school. Finally, it organizes interfaith and intercultural activities to inform the

33

We have some data derived from the experiments that may enable us to gage the impact of the scandal on Irish

Catholics’ generosity and on their attitudes towards their faith and the Church. We have not had time to code and

analyze all that data yet. 34

CITES to Martin report, Cloyne report, and Enda Kenny’s parliamentary speech.

36

French people about Islam and Turkish community. Our interviews were just before the Eid al-

Adha, a major Muslim festival, in which poor are provided with meat. The community was busy

with gathering money to get meat for the Pakistani flood victims.

Alba Intercultural Association, Milan, Italy. The Turkish community in Italy is very small (not

more than 15,000). There are about 4,000 Turks living in Milan, Italy. There is only one Gülen

community association in Milan, Alba Intercultural Association. There are about 200 people who

are somehow affiliated with the association. As in Paris, the association offers weekly religious

education lectures, weekend school for the Turkish immigrant schoolchildren, and

intercultural/interfaith activities. In Milan, the association focuses more on intercultural activities

than on religious education for Muslims. The main target of the association is the Italian people.

In doing so, they aim to inform them about Islam and Turkish culture. Unlike other cities,

weekly Friday prayers are performed in the association. Like Paris, our interviews were just

before the Eid al-Adha, a major Muslim festival, in which poor are provided with meat. The

community was busy with gathering support for the Pakistani flood victims.

Turkish-Irish Educational and Cultural Society, Dublin, Ireland. The Turkish immigrant

community in Ireland is the smallest among all the four cities. An officer from the the Turkish

embassy said that there is about 5,000 Turks living in Ireland, most of which live in Dublin. The

Gülen community has only one association in Dublin, TIECS (Turkish-Irish Educational and

Cultural Society). The association is the smallest of all. There are about 100-125 people

somehow affiliated with the association. They have unorganized weekend classes for the Turkish

schoolchildren. Like Italy, their major concentration is interfaith/intercultural dialogue. The

major concentration during my stay in Dublin was rental of a bigger place for the association,

preferably in the city center. The Gülen community in Ireland also had campaigns for Pakistani

flood victims during the Eid al-Adha festival. right before we conducted our interviews.

37

Works Cited

Abrams, Dominic, Margaret Wetherell, Sandra Cochrane, Michael Hogg, and John Turner. 1990.

“Knowing What to Think by Knowing Who You Are: Self-Categorization and the Nature of

Norm Formation, Conformity and Group Polarization.” British Journal of Social Psychology.

29: 97-119.

Alesina, Alberto, Reza Bagir, and William Easterly. 1999. “Public Goods and Ethnic Divisions.”

The Quarterly Journal of Economics 114: no. 4 (Nov.): 1243-1284.

Al-Ghazzali, Mohammad. 1966. The Mysteries of Almsgiving: Ihya-i Ulum al-Din Book 5.

Beirut: American University of Beirut. Trans. Nabih Amin Faris.

Allport, G. W. and J. M. Ross. 1967. “Personal orientation and prejudice.” Journal of

Personality and Social Psychology 5: 432-443.

Andreoni, James. 1990. “Impure Altruism and Donations to Public Goods: A Theory of Warm-

Glow Giving.” The Economic Journal 100 (June): 464-477

Andreoni, James and J.H. Miller. 2002. "Giving According to GARP: An Experimental Test of

the Consistency of Preferences for Altruism.'' Econometrica, 70/2 (March): 737-753

Andreoni, James and Justin Rao. 2011. “The Power of Asking: How communication Affects

Selfishness, Empathy, and Altruism.” Journal of Public Economics 95: 513-520.

Atran, Scott. 2002. In Gods we Trust: The Evolutionary Landscape of Religion. New York:

Oxford University Press.

Atran, Scott and Ara Norenzayan. 2004. “Religion’s evolutionary landscape: Counterintuition,

commitment, compassion, communion.” Behavioral and Brain Sciences 27: 713-770.

Axelrod, Robert M. 1984. The Evolution of Cooperation. New York: Basic Books.

Axelrod, Robert M. 1997. The Complexity of Cooperation. Agent-based Models of Competition

and Collaboration. Princeton: Princeton University Press.

Bailey, Betty Jane and J. Martin Bailey. 2003. Who Are the Christians in the Middle East? Grand

Rapids, MI: William B. Eerdmans.

Batson, C. Daniel. 2011. Altruism in Humans. Oxford: Oxford University Press.

Bekkers, René and Theo Schuyt. 2007. “And Who is Your Neighbor? Explaining

Denominational Differences in Charitable Giving and Volunteering in the Netherlands” Review

of Religious Research. 50/1 (Sept.): 74-96.

Bendor, Jonathan, Daniel Diermeier, David A. Siegel, Michael M. Ting. 2011. A Behavioral

Theory of Elections. Princeton: Princeton University Press.

Benthall, Jonathan and Jérôme Bellion-Jourdan. 2003. The Charitable Crescent. Politics of Aid

in the Muslim World. London: I.B. Tauris.

Berman, Eli. 2009. Radical, Religious and Violent: The New Economics of Terrorism.

Cambridge, MA: MIT Press.

Berman, Eli, and David D. Laitin. 2008. "Religion, Terrorism, and Public Goods: Testing the

Club Model." Journal of Public Economics 92, no. 10-11 (October): 1942-1967.

Berman, Sheri. 2001. "Ideas, Norms, and Culture in Political Analysis." Comparative Politics

33, no. 2 (January): 231-250.

Boix, Carles and Daniel Posner. 1998. “Social Capital: Explaining Its Origins and Effects on

Governmental Performance.” British Journal of Political Science. 28/4: 686-693

Bowles, Samuel and Herbert Gintis. 2004. “The Evolution of Strong Reciprocity: Cooperation in

Heterogeneous Populations.” Theoretical Population Biology 65: 17-28.

Boyer, Pascal. 2001. Religion Explained: The Evolutionary Origins of Religious Thought. New

York: Basic Books.

38

Buchanan, James. 1965. “An Economic Theory of Clubs.” Economica 32/125: 1-14.

Calhoun-Brown, Allison. 2000. “Upon This Rock: The Black Church, Nonviolence, and the

Civil Rights Movement.” PS: Political Science and Politics 33, No. 2. (Jun.): 168-174.

Caputo, Richard K. 2009. "Religious Capital and Intergenerational Transmission of Volunteering

as Correlates of Civic Engagement." Nonprofit and Voluntary Sector Quarterly 38 (October).

Casari, Marco and Charles R. Plott. 2003. “Decentralized Management of Common Property

Resources: Experiments with a Centuries-Old Institution.” Journal of Economic Behavior &

Organization 51: 217-247.

Chamlee-Wright and Virgil Henry Storr. 2009. “Club Goods and Post-Disaster Community

Return.” Rationality and Society 21/4.

Chaves, Mark. 2010. “Rain Dances in the Dry Season: Overcoming the Religious Congruence

Fallacy” Journal for the Scientific Study of Religion, 49/1: 1-14

Catechism of the Catholic Church. 1999. London: Geoffrey Chapman.

Chen, Daniel L. 2010. “Club Goods and Group Identity: Evidence from Islamic Resurgence

during the Indonesian Financial Crisis.” Journal of Political Economy 118, no. 2: 300 – 354.

Cialdini, R. B., Brown, S. L., Lewis, B. P., Luce, C., & Neuberg, S. L. 1997. “Reintepreting the

empathy-altruism relationship: When one into one equals oneness.” Journal of Personality and

Social Psychology, 73, 481-494.

Clark, Janine A. 2004. Islam, Charity, and Activism. Middle-Class Networks and Social Welfare

in Egypt, Jordan, and Yemen. Bloomington: Indiana University Press

Cohen, Adam B. 2009. “Many forms of culture.” American Psychologist, 64: 194-204

Cornes, Richard and Todd Sandler. 1996. The Theory of Externalities, Pubic Goods and Club

Goods, 2nd Edition. Cambridge University Press.

Dhingra, Pawan H., and Penny Edgell Becker. 2001. "Religious Involvement and Volunteering:

Implications for Civil Society." Sociology of Religion 62, no. 3: 315-335.

Edlin, Aaron, Andrew Gelman, and Noah Kaplan. 2007. "Voting as a Rational Choice: Why and

How People Vote to Improve the Well-Being of Others." Rationality and Society 19 (3): 293-

314.

Fearon, James and David D. Laitin. 1996. “Explaining Inter-Ethnic Cooperation” American

Political Science Review 90/4 (Dec 1996): 715-735.

Finnemore, Martha, and Kathryn Sikkink. 2001. "Taking Stock: The Constructivist Research

Program in International Relations and Comparative Politics." Annual Review of Political

Science 4: 391-416.

Fowler, James H. 2006. "Altruism and Turnout." The Journal of Politics 68/3 (Aug.): 674-683.

Fowler, James H. and Cindy Kam. 2007. “Beyond the Self: Social Identity, Altruism, and

Political Participation” Journal of Politics 69/3 (Aug.): 813-827.

Frazee, Charles A. 1983. Catholics and Sultans. The Church and the Ottoman Empire 1453-

1923. Cambridge: Cambridge University Press.

Gerring, John. 2007. Case Study Research: Principles and Practices. New York: Cambridge

University Press

Gill, Anthony. "Religion and Comparative Politics." 2001. Annual Review of Political Science 4:

117-38.

Gill, Anthony. 1998. Rendering Unto Caesar: The Catholic Church and the State in Latin

America. Chicago: University of Chicago Press.

Green, Donald and Ian Shapiro. 1994. Pathologies of Rational Choice: A Critique of

Applications in Political Science. New Haven, CT: Yale University Press.

39

Greif, Avner. 1994. “Cultural Beliefs and the Organization of Society: A Historical and

Theoretical Reflection on Collectivist and Individualist Societies” the Journal of Political

Economy 102/5 (Oct.): 912-950.

Grofman, Bernard. 1993 “Is Turnout the Paradox that Ate Rational Choice Theory?”

Information, Participation, and Choice. Edited by Bernard Grofman. Ann Arbor: University of

Michigan Press.

Habyarimana, James, Macartan Humphreys, Daniel N. Posner and Jeremy M. Weinstein. 2007.

“Why Does Ethnic Diversity Undermine Public Goods Provision?” American Political Science

Review 101/4 (Nov.): 709-726.

Harris, Fredrick C. 1999. Something Within: Religion in African-American Political Activism.

New York: Oxford University Press.

Henrich, Joseph, Steven J. Heine, and Ara Norenzayan (2010) “The Weirdest People in the

World? Behavioral and Brain Sciences, 1-75.

Hill, P. C., & Pargament, K. I. 2003. “Advances in the conceptualization and measurement of

religion and spirituality. Implications for physical and mental health research.” American

Psychologist 58: 64-74.

Himes, Kenneth R. 2006. “Vatican II and Contemporary Politics.” In The Catholic Church and

the Nation-State. Comparative Perspectives. Eds. Paul Christopher Manuel, Lawrence C.

Reardon, and Clyde Wilcox, Washington D.C.: Georgetown University Press, 15-32.

Hoffman, Elizabeth, Kevin McCabe, and Vernon Smith. 1996. “Social Distance and Other-

Regarding Behavior in Dictator Games.” American Economic Review 86/3: 653-60..

Hurd, Elizabeth Shakman. 2008. The Politics of Secularism in International Relations.

Princeton: Princeton University Press.

Iannaccone, Laurence R. 1992. “Sacrifice and Stigma” Journal of Political Economy 100/2: 271-

291.

Iannaccone, Laurence R. 1998. “Introduction to the Economics of Religion” Journal of

Economic Literature 36/3 (Sept.): 1465-1495.

Iannaccone, Laurence R., and Eli Berman. 2006. "Religious Extremism: The Good, the Bad, and

the Deadly" Public Choice 128: 109-129.

Jankowski, Richard. 2002. “Buying a Lottery Ticket to Help the Poor: Altruism versus Self-

interest in the Decision to Vote” Rationality and Society 14: 55–77.

Jankowski, Richard. 2007. "Altruism and the Decision to Vote: Explaining and Testing High

Voter Turnout." Rationality and Society 19/1: 5-34.

Ji, C. and Y. Ibrahim. 2007. “Islamic doctrinal orthodoxy and religious orientations: Scales

development and validation.” International Journal for the Psychology of Religion, 17: 189-208.

Jones-Correa, Michael A. and David L. Leal. 2001. “Political Participation: Does Religion

Matter?” Political Research Quarterly 54/4 (Dec.): 751-770.

Juergensmeyer, Mark. 2000. Terror in the Mind of God: The Global Rise of Religious Violence.

Berkeley: University of California Press.

Kniss, Fred, and Paul D. Numrich. 2007. Sacred Assemblies and Civic Engagement: How

Religion Matters for America's Newest Immigrants. New Brunswick: Rutgers University Press.

Kolstad, Charles D. 2011. “Public Goods Agreements with Other-Regarding Preferences.”

Working Paper. National Bureau of Economic Research: NBER Working Paper No. 17017. May

2011.

Kozlowski, Gregory C. 1998. “Religious Authority, Reform, and Philanthropy in the

Contemporary Muslim World.” In Philanthropy in the World’s Traditions, eds. Warren F.

40

Ilchman, Stanley N. Katz and Edward L. Queen II. Bloomington: Indiana University Press, 279-

308.

Kuran, Timur. 2001. “The Provision of Public Goods under Islamic Law: Origins, Impact, and

Limitations of the Waqf System.” Law and Society Review 35/4: 841-897.

Kurien, Prema. 2001. “Religion, Ethnicity and Politics: Hindu and Muslim Indian Immigration in

the United States.” Ethnic and Racial Studies 24/2: 263-293.

Kurzban, Robert and Daniel Houser. 2005. “Experiments Investigating Cooperative Types in

Humans: A Complement to Evolutionary Theory and Simulations.” Proceedings of the National

Academy of Sciences of the United States of America 102/5: 1803-1807.

Le Coz, Raymond. 1995. L’Église d’Orient. Chrétiens d’Irak, d’Iran et de Turquie. Paris: Les

Éditions du Cerf.

Levine, David K. and Thomas R. Palfrey. 2007. “The Paradox of Voter Participation? A

Laboratory Study.” The American Political Science Review 101/1: 143-158.

Levitt, Matthew and Dennis Ross. 2006. Hamas: Politics, Charity, and Terrorism in the Service

of Jihad. New Haven: Yale University Press.

Lichbach, Mark Irving. 1998. The Rebel’s Dilemma. University of Michigan Press.

Loveland, Matthew T., David Sikkink, Daniel J. Myers, and Benjamin Radcliff. 2005. "Private

Prayer and Civic Involvement." Journal for the Scientific Study of Religion 44, no. 1: 1-14.

Luttmer, Erzo F.P. 2001. “Group Loyalty and the Taste for Redistribution” Journal of Political

Economy. 109: 500-528

Maclean, A. Michael, Lawrence J. Walker and M. Kyle Matsuba. 2004. “Transcendence and the

Moral Self: Identity, Integration, Religion and Moral Life.” Journal for the Scientific Study of

Religion. 43/3: 429-437.

Makowsky, Michael D. 2011. “A Theory of Liberal Churches.” Mathematical Social Sciences

61: 41–51

Mauss, Marcel. 1954/1967. The Gift: forms and functions of exchange in archaic societies. trans.

by Ian Cunnison. New York: Norton.

McAdam, Doug. 1982. Political Process and the Development of Black Insurgency, 1930-1970.

Chicago: The University of Chicago Press.

McBride, Michael. 2007. “Club Mormon: Free-Riders, Monitoring, and Exclusion in the LDS

Church” Rationality and Society 19/4: 395-424.

McBrien, Richard P. 1980. Catholicism. Vol. II. Minneapolis, MN: Winston Press.

McVeigh, Rory and David Sikkink. 2001. “God, Politics, and Protest: Religious Beliefs and the

Legitimation of Contentious Tactics.” Social Forces 79(4): 1425-58.

Mitchell, Joshua. 2007. "Religion is Not a Preference." The Journal of Politics 69, no. 2 (May):

351-362.

Morris, Aldon. 1981. "Black Southern Sit-in Movement: An Analysis of Internal Organization."

American Sociological Review. 46:744-67.

Norenzayan, Ara and Azim F. Shariff. 2008. “The Origin and Evolution of Religious

Prosociality.” Science 322 (Oct.): 58-62.

Norris, Pippa and Ronald Inglehart. 2004. Sacred and Secular: Religion and Politics Worldwide.

Cambridge: Cambridge University Press.

Olson, Mancur. 1965. The Logic of Collective Action. Cambridge, MA: Harvard University

Press.

Orman, Wafa Hakim. 2011. “Survival of the Strictest? An Experimental Examination of Club

Goods.” Working Paper, March 24.

41

Peterson, Nicolas. 1993. “Demand Sharing: Reciprocity and the Pressure for Generosity among

Foragers” American Anthropologist 95/4: 860-874.

Philpott, Daniel. 2000. "The Religious Roots of Modern International Relations." World Politics

52, no. 2 (January): 206-245.

Philpott, Daniel. 2007. "Explaining the Political Ambivalence of Religion." The American

Political Science Review 101, no. 3 (August): 505-525.

Pope Benedict XVI. 2009. Encyclical Letter Caritas in Veritate of the Supreme Pontiff Benedict

XVI. Vatican City.

Popkin, Samuel J. 1979. The Rational Peasant: The Political Economy of Rural Society in

Vietnam. Berkeley: University of California Press.

Putnam, Robert D. 2007. “E Pluribus Unum: Diversity and Community in the Twenty-First

Century.” Scandinavian Political Studies 30, No. 2: 137-174.

Putnam, Robert D. 2000. Bowling Alone: The Collapse and Revival of American Community.

New York: Simon and Schuster.

Queen, Edward L., II. 1996. “The Religious Roots of Philanthropy in the West: Judaism,

Christianity and Islam.” Bloomington, IN: Indiana University Center on Philanthropy. Working

Paper No. 96-4.

Qur’an, The. 2004. trans. M.A.S. Abdel Haleem. New York: Oxford University Press.

Rhodes, R.A.W., Sarah A. Binder and Bert A. Rockman, eds. 2006. The Oxford Handbook of

Political Institutions. Oxford: Oxford University Press.

Rozin, Paul. 2003. Five potential principles for understanding cultural differences in relation to

individual differences. Journal of Research in Personality, 37, 273-283.

Ruane, Joseph and Jennifer Todd. 2011. “Ethnicity and Religion.” Routledge Handbook of

Ethnic Conflict. Edited by Karl Cordell and Stefan Wolff. New York: Routledge.

Sahlins, Marshall. 1972. Stone Age Economics. Chicago: Aldine.

Samuelson, Paul. 1954. “The Pure Theory of Public Expenditure.” Review of Economics and

Statistics 36: 387-89.

Segal, Uzi and Joel Sobel. 2007. "Tit for tat: Foundations of Prestigious Preferences for

Reciprocity in Strategic Settings." Journal of Economic Theory 136: 197-216.

Shariff, Azim F. and Ara Norenzayan. 2007. “God is Watching You. Priming God concepts

Increases Prosocial Behavior in an Anonymous Economic Game.” Psychological Science 18/9:

803-809.

Shayo, Moses. 2009. “A Model of Social Identity with an Application to Political Economy:

Nation, Class, and Redistribution.” American Political Science Review. 103/2 (May): 147-174.

Simpson, Brent and Robb Willer. 2008. “Altruism and Indirect Reciprocity: The Interaction of

Person and Situation in Prosocial Behavior.” Social Psychology Quarterly 71(1): 37-52.

Smith, Kevin B. 2006. “Representational Altruism: The Wary Cooperator as Authoritative

Decision Maker.” American Journal of Political Science 50/4 (Oct.): 1013-1022.

Soetevent, Adriaan R. 2005. “Anonymity in giving in a natural context—a field experiment in 30

churches” Journal of Public Economics 89: 2301-2323.

Tajfel, H. and J. C. Turner. 1979. “An integrative theory of intergroup conflict.” In Social

Psychology of Intergroup Relations, eds. W. G. Austin & S. Worchel. Monterey, CA: Brooks-

Cole.

Tarrow, Sidney. 1994. Power in Movement: Social Movements and Contentious Politics.

Revised and Updated Third Edition. Cambridge: Cambridge University Press.

Tcholakian, Hovhannes J. 1998. L’Eglise Armenienne Catholique en Turquie. Istanbul: Ohan

42

Matbaacilik.

Thomas, Scott. 2005. The Global Resurgence of Religion and the Transformation of

International Relations. New York: Palgrave MacMillan.

Trejo, Guillermo. 2009. "Religious Competition and Ethnic Mobilization in Latin America: Why

the Catholic Church Promotes Indigenous Movements in Mexico." American Political Science

Review 103, no. 3 (August): 323-342.

Verba, Sidney, Kay Lehman Schlozman, and Henry E. Brady. 1995. Voice and Equality: Civic

Voluntarism in American Politics. Cambridge: Harvard University Press.

Warner, Carolyn M. 2000. Confessions of an Interest Group. Princeton: Princeton University

Press.

Warneken, Felix, et. al. 2006. “Altruistic Helping in Human Infants and Young Chimpanzees.”

Science 311: 1301-3

Warneken, Felix and Michael Tomasello. 2008. “Extrinsic rewards undermine altruistic

tendencies in 20 month-olds.” Developmental Psychology 44/6: 1785-1788.

Warneken, Felix and Michael Tomasello. 2009. “The Roots of human altruism.” British Journal

of Psychology 100: 455-471

Wiktorowicz, Quintan. 2004. “Islamic Activism and Social Movement Theory.” in Islamic

Activism: A Social Movement Theory Approach.” Edited by Quintan Wiktorowicz:

Bloomington: Indiana University Press.

Will, Jeffry A. and John K. Cochran. 1995. “God Helps Those Who Help Themselves?: The

Effects of Religious Affiliation, Religiosity, and Deservedness on Generosity Toward the Poor.”

Sociology of Religion 56/3: 327-338.

Wuthnow, Robert. 1991. Acts of Compassion. Caring for Others and Helping Ourselves.

Princeton: Princeton University Press.

Wuthnow, Robert J. 2011. “Taking Talk Seriously: Religious Discourse as Social Practice”

Journal for the Scientific Study of Religion 50/1 (March): 1-21.